title
stringlengths 6
244
| abstract
stringlengths 19
5.09k
| label
stringclasses 10
values |
|---|---|---|
Regularizing Double Machine Learning in Partially Linear Endogenous
Models | We estimate the linear coefficient in a partially linear model with
confounding variables. We rely on double machine learning (DML) and extend it
with an additional regularization and selection scheme. We allow for more
general dependence structures among the model variables than what has been
investigated previously, and we prove that this DML estimator remains
asymptotically Gaussian and converges at the parametric rate. The DML estimator
has a two-stage least squares interpretation and may produce overly wide
confidence intervals. To address this issue, we propose the
regularization-selection regsDML method that leads to narrower confidence
intervals. It is fully data driven and optimizes an estimated asymptotic mean
squared error of the coefficient estimate. Empirical examples demonstrate our
methodological and theoretical developments. Software code for our regsDML
method will be made available in the R-package dmlalg.
| stat |
Alleviating confounding in spatio-temporal areal models with an
application on crimes against women in India | Assessing associations between a response of interest and a set of covariates
in spatial areal models is the leitmotiv of ecological regression. However, the
presence of spatially correlated random effects can mask or even bias estimates
of such associations due to confounding effects if they are not carefully
handled. Though potentially harmful, confounding issues have often been ignored
in practice leading to wrong conclusions about the underlying associations
between the response and the covariates. In spatio-temporal areal models, the
temporal dimension may emerge as a new source of confounding, and the problem
may be even worse. In this work, we propose two approaches to deal with
confounding of fixed effects by spatial and temporal random effects, while
obtaining good model predictions. In particular, restricted regression and an
apparently -- though in fact not -- equivalent procedure using constraints are
proposed within both fully Bayes and empirical Bayes approaches. The methods
are compared in terms of fixed-effect estimates and model selection criteria.
The techniques are used to assess the association between dowry deaths and
certain socio-demographic covariates in the districts of Uttar Pradesh, India.
| stat |
Route Identification in the National Football League | Tracking data in the NFL is a sequence of spatial-temporal measurements that
vary in length depending on the duration of the play. In this paper, we
demonstrate how model-based curve clustering of observed player trajectories
can be used to identify the routes run by eligible receivers on offensive
passing plays. We use a Bernstein polynomial basis function to represent
cluster centers, and the Expectation Maximization algorithm to learn the route
labels for each of the 34,698 routes run on the 6,963 passing plays in the data
set. We go on to suggest ideas for new potential receiver metrics that account
for receiver deployment. The resulting route labels can also be paired with
film to enable streamlined queries of game film.
| stat |
Adaptive Non-Parametric Regression With the $K$-NN Fused Lasso | The fused lasso, also known as total-variation denoising, is a
locally-adaptive function estimator over a regular grid of design points. In
this paper, we extend the fused lasso to settings in which the points do not
occur on a regular grid, leading to an approach for non-parametric regression.
This approach, which we call the $K$-nearest neighbors ($K$-NN) fused lasso,
involves (i) computing the $K$-NN graph of the design points; and (ii)
performing the fused lasso over this $K$-NN graph. We show that this procedure
has a number of theoretical advantages over competing approaches: specifically,
it inherits local adaptivity from its connection to the fused lasso, and it
inherits manifold adaptivity from its connection to the $K$-NN approach. We
show that excellent results are obtained in a simulation study and on an
application to flu data. For completeness, we also study an estimator that
makes use of an $\epsilon$-graph rather than a $K$-NN graph, and contrast this
with the $K$-NN fused lasso.
| stat |
Assessing and relaxing the Markov assumption in the illness-death model | Multi-state survival analysis considers several potential events of interest
along a disease pathway. Such analyses are crucial to model complex patient
trajectories and are increasingly being used in epidemiological and health
economic settings. Multi-state models often make the Markov assumption, whereby
an individual's future trajectory is dependent only upon their present state,
not their past. In reality, there may be transitional dependence upon either
previous events and/or more than one timescale, for example time since entry to
the current or previous state(s). The aim of this study was to develop an
illness-death Weibull model allowing for multiple timescales to impact the
future risk of death. Following this, we evaluated the performance of the
multiple timescale model against a Markov illness-death model in a set of
plausible simulation scenarios when the Markov assumption was violated. Guided
by a study in breast cancer, data were simulated from Weibull baseline
distributions, with hazard functions dependent on single and multiple
timescales. Markov and non-Markov models were fitted to account for/ignore the
underlying data structure. Ignoring the presence of multiple timescales led to
bias in underlying transition rates between states and associated covariate
effects, while transition probabilities and lengths of stay were fairly
robustly estimated. Further work may be needed to evaluate different estimands
or more complex multi-state models. Software implementations in Stata are also
described for simulating and estimating multiple timescale multi-state models.
| stat |
Probabilistic Temperature Forecasting with a Heteroscedastic
Autoregressive Ensemble Postprocessing model | Weather prediction today is performed with numerical weather prediction (NWP)
models. These are deterministic simulation models describing the dynamics of
the atmosphere, and evolving the current conditions forward in time to obtain a
prediction for future atmospheric states. To account for uncertainty in NWP
models it has become common practice to employ ensembles of NWP forecasts.
However, NWP ensembles often exhibit forecast biases and dispersion errors,
thus require statistical postprocessing to improve reliability of the ensemble
forecasts. This work proposes an extension of a recently developed
postprocessing model utilizing autoregressive information present in the
forecast error of the raw ensemble members. The original approach is modified
to let the variance parameter depend on the ensemble spread, yielding a
two-fold heteroscedastic model. Furthermore, an additional high-resolution
forecast is included into the postprocessing model, yielding improved
predictive performance. Finally, it is outlined how the autoregressive model
can be utilized to postprocess ensemble forecasts with higher forecast
horizons, without the necessity of making fundamental changes to the original
model. We accompany the new methodology by an implementation within the R
package ensAR to make our method available for other researchers working in
this area. To illustrate the performance of the heteroscedastic extension of
the autoregressive model, and its use for higher forecast horizons we present a
case study for a data set containing 12 years of temperature forecasts and
observations over Germany. The case study indicates that the autoregressive
model yields particularly strong improvements for forecast horizons beyond 24
hours.
| stat |
On quantile oriented sensitivity analysis | We propose to study quantile oriented sensitivity indices (QOSA indices) and
quantile oriented Shapley effects (QOSE). Some theoretical properties of QOSA
indices will be given and several calculations of QOSA indices and QOSE will
allow to better understand the behaviour and the interest of these indices.
| stat |
The statistical finite element method (statFEM) for coherent synthesis
of observation data and model predictions | The increased availability of observation data from engineering systems in
operation poses the question of how to incorporate this data into finite
element models. To this end, we propose a novel statistical construction of the
finite element method that provides the means of synthesising measurement data
and finite element models. The Bayesian statistical framework is adopted to
treat all the uncertainties present in the data, the mathematical model and its
finite element discretisation. From the outset, we postulate a data-generating
model which additively decomposes data into a finite element, a model
misspecification and a noise component. Each of the components may be uncertain
and is considered as a random variable with a respective prior probability
density. The prior of the finite element component is given by a conventional
stochastic forward problem. The prior probabilities of the model
misspecification and measurement noise, without loss of generality, are assumed
to have zero-mean and known covariance structure. Our proposed statistical
model is hierarchical in the sense that each of the three random components may
depend on non-observable random hyperparameters. Because of the hierarchical
structure of the statistical model, Bayes rule is applied on three different
levels in turn to infer the posterior densities of the three random components
and hyperparameters. On level one, we determine the posterior densities of the
finite element component and the true system response using the prior finite
element density given by the forward problem and the data likelihood. On the
next level, we infer the hyperparameter posterior densities from their
respective priors and the marginal likelihood of the first inference problem.
Finally, on level three we use Bayes rule to choose the most suitable finite
element model in light of the observed data by computing the model posteriors.
| stat |
Flattening the COVID-19 Curve: The "Greek" case in the Global Pandemic | The global crisis caused by the COVID-19 pandemic, in conjunction with the
economic consequences and the collapse of health systems, has raised serious
concerns in Europe, which is the most affected continent by the pandemic since
it recorded 2,388,694 cases and 190,091 deaths (39.6% of the worldwide total),
of which 71.7% (136,238) are in the United Kingdom (43,414), Italy (34,708),
France (29,778), and Spain (28,338). Unlike other countries, Greece, with about
310 confirmed cases and 18 deaths per million, is one bright exception in the
study and analysis of this phenomenon. Focusing on the peculiarities of the
disease spreading in Greece, both in epidemiological and in implementation
terms, this paper applies an exploratory analysis of COVID-19 temporal spread
in Greece and proposes a methodological approach for the modeling and
prediction of the disease based on the Regression Splines algorithm and the
change rate of the total infections. Also, it proposes a hybrid spline
regression and complex network model of social distance measures evaluating and
interpreting the spread of the disease. The overall approach contributes to
decision making and support of the public health system and to the fight
against the pandemic.
| stat |
Proximal Langevin Algorithm: Rapid Convergence Under Isoperimetry | We study the Proximal Langevin Algorithm (PLA) for sampling from a
probability distribution $\nu = e^{-f}$ on $\mathbb{R}^n$ under isoperimetry.
We prove a convergence guarantee for PLA in Kullback-Leibler (KL) divergence
when $\nu$ satisfies log-Sobolev inequality (LSI) and $f$ has bounded second
and third derivatives. This improves on the result for the Unadjusted Langevin
Algorithm (ULA), and matches the fastest known rate for sampling under LSI
(without Metropolis filter) with a better dependence on the LSI constant. We
also prove convergence guarantees for PLA in R\'enyi divergence of order $q >
1$ when the biased limit satisfies either LSI or Poincar\'e inequality.
| stat |
Google Street View image of a house predicts car accident risk of its
resident | Road traffic injuries are a leading cause of death worldwide. Proper
estimation of car accident risk is critical for appropriate allocation of
resources in healthcare, insurance, civil engineering, and other industries. We
show how images of houses are predictive of car accidents. We analyze 20,000
addresses of insurance company clients, collect a corresponding house image
using Google Street View, and annotate house features such as age, type, and
condition. We find that this information substantially improves car accident
risk prediction compared to the state-of-the-art risk model of the insurance
company and could be used for price discrimination. From this perspective,
public availability of house images raises legal and social concerns, as they
can be a proxy of ethnicity, religion and other sensitive data.
| stat |
Robust Estimation and Variable Selection for the Accelerated Failure
Time Model | This paper considers robust modeling of the survival time for cancer
patients. Accurate prediction can be helpful for developing therapeutic and
care strategies. We propose a unified Expectation-Maximization approach
combined with the L1-norm penalty to perform variable selection and obtain
parameter estimation simultaneously for the accelerated failure time model with
right-censored survival data. Our approach can be used with general loss
functions, and reduces to the well-known Buckley-James method when the
squared-error loss is used without regularization. To mitigate the effects of
outliers and heavy-tailed noise in the real application, we advocate the use of
robust loss functions under our proposed framework. Simulation studies are
conducted to evaluate the performance of the proposed approach with different
loss functions, and an application to an ovarian carcinoma study is provided.
Meanwhile, we extend our approach by incorporating the group structure of
covariates.
| stat |
Guaranteed Simultaneous Asymmetric Tensor Decomposition via
Orthogonalized Alternating Least Squares | Tensor CANDECOMP/PARAFAC (CP) decomposition is an important tool that solves
a wide class of machine learning problems. Existing popular approaches recover
components one by one, not necessarily in the order of larger components first.
Recently developed simultaneous power method obtains only a high probability
recovery of top $r$ components even when the observed tensor is noiseless. We
propose a Slicing Initialized Alternating Subspace Iteration (s-ASI) method
that is guaranteed to recover top $r$ components ($\epsilon$-close)
simultaneously for (a)symmetric tensors almost surely under the noiseless case
(with high probability for a bounded noise) using $O(\log(\log
\frac{1}{\epsilon}))$ steps of tensor subspace iterations. Our s-ASI method
introduces a Slice-Based Initialization that runs
$O(1/\log(\frac{\lambda_r}{\lambda_{r+1}}))$ steps of matrix subspace
iterations, where $\lambda_r$ denotes the r-th top singular value of the
tensor. We are the first to provide a theoretical guarantee on simultaneous
orthogonal asymmetric tensor decomposition. Under the noiseless case, we are
the first to provide an \emph{almost sure} theoretical guarantee on
simultaneous orthogonal tensor decomposition. When tensor is noisy, our
algorithm for asymmetric tensor is robust to noise smaller than
$\min\{O(\frac{(\lambda_r - \lambda_{r+1})\epsilon}{\sqrt{r}}),
O(\delta_0\frac{\lambda_r -\lambda_{r+1}}{\sqrt{d}})\}$, where $\delta_0$ is a
small constant proportional to the probability of bad initializations in the
noisy setting.
| stat |
Quantum Compressed Sensing with Unsupervised Tensor-Network Machine
Learning | We propose tensor-network compressed sensing (TNCS) by combining the ideas of
compressed sensing, tensor network (TN), and machine learning, which permits
novel and efficient quantum communications of realistic data. The strategy is
to use the unsupervised TN machine learning algorithm to obtain the entangled
state $|\Psi \rangle$ that describes the probability distribution of a huge
amount of classical information considered to be communicated. To transfer a
specific piece of information with $|\Psi \rangle$, our proposal is to encode
such information in the separable state with the minimal distance to the
measured state $|\Phi \rangle$ that is obtained by partially measuring on
$|\Psi \rangle$ in a designed way. To this end, a measuring protocol analogous
to the compressed sensing with neural-network machine learning is suggested,
where the measurements are designed to minimize uncertainty of information from
the probability distribution given by $|\Phi \rangle$. In this way, those who
have $|\Phi \rangle$ can reliably access the information by simply measuring on
$|\Phi \rangle$. We propose q-sparsity to characterize the sparsity of quantum
states and the efficiency of the quantum communications by TNCS. The high
q-sparsity is essentially due to the fact that the TN states describing nicely
the probability distribution obey the area law of entanglement entropy. Testing
on realistic datasets (hand-written digits and fashion images), TNCS is shown
to possess high efficiency and accuracy, where the security of communications
is guaranteed by the fundamental quantum principles.
| stat |
On the correlation analysis of illiquid stocks | The serial correlations of illiquid stock's price movements are studied, by
taking into account for the effect of the stochastic zero returns in various
situations. More precisely, heteroscedasticity and time-varying zero returns
probability patterns are considered. Depending on the set up, we investigate
how the usual autocorrelations can be accommodated, to deliver an accurate
representation of the price changes serial correlations. An index and a
cumulative sum (CUMSUM) test for choosing the adequate tool are provided. We
shed some light on the properties of the different serial correlations
measures, by mean of Monte Carlo experiments. The theoretical arguments are
illustrated considering shares from the Chilean stock market.
| stat |
Machine Learning Econometrics: Bayesian algorithms and methods | As the amount of economic and other data generated worldwide increases
vastly, a challenge for future generations of econometricians will be to master
efficient algorithms for inference in empirical models with large information
sets. This Chapter provides a review of popular estimation algorithms for
Bayesian inference in econometrics and surveys alternative algorithms developed
in machine learning and computing science that allow for efficient computation
in high-dimensional settings. The focus is on scalability and parallelizability
of each algorithm, as well as their ability to be adopted in various empirical
settings in economics and finance.
| stat |
Central limit theorems for directional and linear random variables with
applications | A central limit theorem for the integrated squared error of the
directional-linear kernel density estimator is established. The result enables
the construction and analysis of two testing procedures based on squared loss:
a nonparametric independence test for directional and linear random variables
and a goodness-of-fit test for parametric families of directional-linear
densities. Limit distributions for both test statistics, and a consistent
bootstrap strategy for the goodness-of-fit test, are developed for the
directional-linear case and adapted to the directional-directional setting.
Finite sample performance for the goodness-of-fit test is illustrated in a
simulation study. This test is also applied to datasets from biology and
environmental sciences.
| stat |
An ARIMA model to forecast the spread and the final size of COVID-2019
epidemic in Italy | Coronavirus disease (COVID-2019) is a severe ongoing novel pandemic that is
spreading quickly across the world. Italy, that is widely considered one of the
main epicenters of the pandemic, has registered the highest COVID-2019 death
rates and death toll in the world, to the present day. In this article I
estimate an autoregressive integrated moving average (ARIMA) model to forecast
the epidemic trend over the period after April 4, 2020, by using the Italian
epidemiological data at national and regional level. The data refer to the
number of daily confirmed cases officially registered by the Italian Ministry
of Health (www.salute.gov.it) for the period February 20 to April 4, 2020. The
main advantage of this model is that it is easy to manage and fit. Moreover, it
may give a first understanding of the basic trends, by suggesting the
hypothetic epidemic's inflection point and final size.
| stat |
MP-Boost: Minipatch Boosting via Adaptive Feature and Observation
Sampling | Boosting methods are among the best general-purpose and off-the-shelf machine
learning approaches, gaining widespread popularity. In this paper, we seek to
develop a boosting method that yields comparable accuracy to popular AdaBoost
and gradient boosting methods, yet is faster computationally and whose solution
is more interpretable. We achieve this by developing MP-Boost, an algorithm
loosely based on AdaBoost that learns by adaptively selecting small subsets of
instances and features, or what we term minipatches (MP), at each iteration. By
sequentially learning on tiny subsets of the data, our approach is
computationally faster than other classic boosting algorithms. Also as it
progresses, MP-Boost adaptively learns a probability distribution on the
features and instances that upweight the most important features and
challenging instances, hence adaptively selecting the most relevant minipatches
for learning. These learned probability distributions also aid in
interpretation of our method. We empirically demonstrate the interpretability,
comparative accuracy, and computational time of our approach on a variety of
binary classification tasks.
| stat |
Compressive Learning for Semi-Parametric Models | In the compressive learning theory, instead of solving a statistical learning
problem from the input data, a so-called sketch is computed from the data prior
to learning. The sketch has to capture enough information to solve the problem
directly from it, allowing to discard the dataset from the memory. This is
useful when dealing with large datasets as the size of the sketch does not
scale with the size of the database. In this paper, we reformulate the original
compressive learning framework to explicitly cater for the class of
semi-parametric models. The reformulation takes account of the inherent
topology and structure of semi-parametric models, creating an intuitive pathway
to the development of compressive learning algorithms. We apply our developed
framework to both the semi-parametric models of independent component analysis
and subspace clustering, demonstrating the robustness of the framework to
explicitly show when a compression in complexity can be achieved.
| stat |
Prior Adaptive Semi-supervised Learning with Application to EHR
Phenotyping | Electronic Health Records (EHR) data, a rich source for biomedical research,
have been successfully used to gain novel insight into a wide range of
diseases. Despite its potential, EHR is currently underutilized for discovery
research due to it's major limitation in the lack of precise phenotype
information. To overcome such difficulties, recent efforts have been devoted to
developing supervised algorithms to accurately predict phenotypes based on
relatively small training datasets with gold standard labels extracted via
chart review. However, supervised methods typically require a sizable training
set to yield generalizable algorithms especially when the number of candidate
features, $p$, is large. In this paper, we propose a semi-supervised (SS) EHR
phenotyping method that borrows information from both a small labeled data
where both the label $Y$ and the feature set $X$ are observed and a much larger
unlabeled data with observations on $X$ only as well as a surrogate variable
$S$ that is predictive of $Y$ and available for all patients, under a high
dimensional setting. Under a working prior assumption that $S$ is related to
$X$ only through $Y$ and allowing it to hold approximately, we propose a prior
adaptive semi-supervised (PASS) estimator that adaptively incorporates the
prior knowledge by shrinking the estimator towards a direction derived under
the prior. We derive asymptotic theory for the proposed estimator and
demonstrate its superiority over existing estimators via simulation studies.
The proposed method is applied to an EHR phenotyping study of rheumatoid
arthritis at Partner's Healthcare.
| stat |
Anomaly Detection on Seasonal Metrics via Robust Time Series
Decomposition | The stability and persistence of web services are important to Internet
companies to improve user experience and business performances. To keep eyes on
numerous metrics and report abnormal situations, time series anomaly detection
methods are developed and applied by various departments in companies and
institutions. In this paper, we proposed a robust anomaly detection algorithm
(MEDIFF) to monitor online business metrics in real time. Specifically, a
decomposition method using robust statistical metric--median--of the time
series was applied to decouple the trend and seasonal components. With the
effects of daylight saving time (DST) shift and holidays, corresponding
components were decomposed from the time series. The residual after
decomposition was tested by a generalized statistics method to detect outliers
in the time series. We compared the proposed MEDIFF algorithm with two open
source algorithms (SH-ESD and DONUT) by using our labeled internal business
metrics. The results demonstrated the effectiveness of the proposed MEDIFF
algorithm.
| stat |
Infinitely deep neural networks as diffusion processes | When the parameters are independently and identically distributed
(initialized) neural networks exhibit undesirable properties that emerge as the
number of layers increases, e.g. a vanishing dependency on the input and a
concentration on restrictive families of functions including constant
functions. We consider parameter distributions that shrink as the number of
layers increases in order to recover well-behaved stochastic processes in the
limit of infinite depth. This leads to set forth a link between infinitely deep
residual networks and solutions to stochastic differential equations, i.e.
diffusion processes. We show that these limiting processes do not suffer from
the aforementioned issues and investigate their properties.
| stat |
Fast Computation of Genome-Metagenome Interaction Effects | Motivation. Association studies have been widely used to search for
associations between common genetic variants observations and a given
phenotype. However, it is now generally accepted that genes and environment
must be examined jointly when estimating phenotypic variance. In this work we
consider two types of biological markers: genotypic markers, which characterize
an observation in terms of inherited genetic information, and metagenomic
marker which are related to the environment. Both types of markers are
available in their millions and can be used to characterize any observation
uniquely. Objective. Our focus is on detecting interactions between groups of
genetic and metagenomic markers in order to gain a better understanding of the
complex relationship between environment and genome in the expression of a
given phenotype. Contributions. We propose a novel approach for efficiently
detecting interactions between complementary datasets in a high-dimensional
setting with a reduced computational cost. The method, named SICOMORE, reduces
the dimension of the search space by selecting a subset of supervariables in
the two complementary datasets. These supervariables are given by a weighted
group structure defined on sets of variables at different scales. A Lasso
selection is then applied on each type of supervariable to obtain a subset of
potential interactions that will be explored via linear model testing. Results.
We compare SICOMORE with other approaches in simulations, with varying sample
sizes, noise, and numbers of true interactions. SICOMORE exhibits convincing
results in terms of recall, as well as competitive performances with respect to
running time. The method is also used to detect interaction between genomic
markers in Medicago truncatula and metagenomic markers in its rhizosphere
bacterial community. Software availability. A R package is available, along
with its documentation and associated scripts, allowing the reader to reproduce
the results presented in the paper.
| stat |
A Semiparametric Approach to Model Effect Modification | One fundamental statistical question for research areas such as precision
medicine and health disparity is about discovering effect modification of
treatment or exposure by observed covariates. We propose a semiparametric
framework for identifying such effect modification. Instead of using the
traditional outcome models, we directly posit semiparametric models on
contrasts, or expected differences of the outcome under different treatment
choices or exposures. Through semiparametric estimation theory, all valid
estimating equations, including the efficient scores, are derived. Besides
doubly robust loss functions, our approach also enables dimension reduction in
presence of many covariates. The asymptotic and non-asymptotic properties of
the proposed methods are explored via a unified statistical and algorithmic
analysis. Comparison with existing methods in both simulation and real data
analysis demonstrates the superiority of our estimators especially for an
efficiency improved version.
| stat |
A generalized linear joint trained framework for semi-supervised
learning of sparse features | The elastic-net is among the most widely used types of regularization
algorithms, commonly associated with the problem of supervised generalized
linear model estimation via penalized maximum likelihood. Its nice properties
originate from a combination of $\ell_1$ and $\ell_2$ norms, which endow this
method with the ability to select variables taking into account the
correlations between them. In the last few years, semi-supervised approaches,
that use both labeled and unlabeled data, have become an important component in
the statistical research. Despite this interest, however, few researches have
investigated semi-supervised elastic-net extensions. This paper introduces a
novel solution for semi-supervised learning of sparse features in the context
of generalized linear model estimation: the generalized semi-supervised
elastic-net (s2net), which extends the supervised elastic-net method, with a
general mathematical formulation that covers, but is not limited to, both
regression and classification problems. We develop a flexible and fast
implementation for s2net in R, and its advantages are illustrated using both
real and synthetic data sets.
| stat |
Compositional uncertainty in deep Gaussian processes | Gaussian processes (GPs) are nonparametric priors over functions. Fitting a
GP implies computing a posterior distribution of functions consistent with the
observed data. Similarly, deep Gaussian processes (DGPs) should allow us to
compute a posterior distribution of compositions of multiple functions giving
rise to the observations. However, exact Bayesian inference is intractable for
DGPs, motivating the use of various approximations. We show that the
application of simplifying mean-field assumptions across the hierarchy leads to
the layers of a DGP collapsing to near-deterministic transformations. We argue
that such an inference scheme is suboptimal, not taking advantage of the
potential of the model to discover the compositional structure in the data. To
address this issue, we examine alternative variational inference schemes
allowing for dependencies across different layers and discuss their advantages
and limitations.
| stat |
A Note on Particle Gibbs Method and its Extensions and Variants | High-dimensional state trajectories of state-space models pose challenges for
Bayesian inference. Particle Gibbs (PG) methods have been widely used to sample
from the posterior of a state space model. Basically, particle Gibbs is a
Particle Markov Chain Monte Carlo (PMCMC) algorithm that mimics the Gibbs
sampler by drawing model parameters and states from their conditional
distributions.
This tutorial provides an introductory view on Particle Gibbs (PG) method and
its extensions and variants, and illustrates through several examples of
inference in non-linear state space models (SSMs). We also implement PG
Samplers in two different programming languages: Python and Rust. Comparison of
run-time performance of Python and Rust programs are also provided for various
PG methods.
| stat |
Interpretable Machines: Constructing Valid Prediction Intervals with
Random Forests | An important issue when using Machine Learning algorithms in recent research
is the lack of interpretability. Although these algorithms provide accurate
point predictions for various learning problems, uncertainty estimates
connected with point predictions are rather sparse. A contribution to this gap
for the Random Forest Regression Learner is presented here. Based on its
Out-of-Bag procedure, several parametric and non-parametric prediction
intervals are provided for Random Forest point predictions and theoretical
guarantees for its correct coverage probability is delivered. In a second part,
a thorough investigation through Monte-Carlo simulation is conducted evaluating
the performance of the proposed methods from three aspects: (i) Analyzing the
correct coverage rate of the proposed prediction intervals, (ii) Inspecting
interval width and (iii) Verifying the competitiveness of the proposed
intervals with existing methods. The simulation yields that the proposed
prediction intervals are robust towards non-normal residual distributions and
are competitive by providing correct coverage rates and comparably narrow
interval lengths, even for comparably small samples.
| stat |
Forecasting Time Series for Power Consumption Data in Different
Buildings Using the Fractional Brownian Motion | In the present paper will be discussed the problem related to the individual
household electric power consumption of objects in different areas-industry,
farmers, banks, hospitals, theaters, hostels, supermarkets, universities. The
main goal of the directed research is to estimate the active P and full S power
consumptions for all studied buildings. The defined goal is achieved by solving
of the following three problems. The first problem studies which buildings
increase their power consumption. The second one finds which objects have the
greatest increase of power consumption. And the third problem regards if it is
possible to make a short-term forecast, based on the solutions of previous two
problems. The present research and solving of the aforementioned problems is
conducted using fractional Brownian motion theory. The applicability of this
approach is illustrated on the example with 20 real objects in different areas.
The paper ends with conclusion notes about possibilities to make short-term
forecasts about power consumption of the considered buildings.
| stat |
Parametrization and generation of geological models with generative
adversarial networks | One of the main challenges in the parametrization of geological models is the
ability to capture complex geological structures often observed in the
subsurface. In recent years, generative adversarial networks (GAN) were
proposed as an efficient method for the generation and parametrization of
complex data, showing state-of-the-art performances in challenging computer
vision tasks such as reproducing natural images (handwritten digits, human
faces, etc.). In this work, we study the application of Wasserstein GAN for the
parametrization of geological models. The effectiveness of the method is
assessed for uncertainty propagation tasks using several test cases involving
different permeability patterns and subsurface flow problems. Results show that
GANs are able to generate samples that preserve the multipoint statistical
features of the geological models both visually and quantitatively. The
generated samples reproduce both the geological structures and the flow
statistics of the reference geology.
| stat |
Do RNN and LSTM have Long Memory? | The LSTM network was proposed to overcome the difficulty in learning
long-term dependence, and has made significant advancements in applications.
With its success and drawbacks in mind, this paper raises the question - do RNN
and LSTM have long memory? We answer it partially by proving that RNN and LSTM
do not have long memory from a statistical perspective. A new definition for
long memory networks is further introduced, and it requires the model weights
to decay at a polynomial rate. To verify our theory, we convert RNN and LSTM
into long memory networks by making a minimal modification, and their
superiority is illustrated in modeling long-term dependence of various
datasets.
| stat |
Fast Multivariate Probit Estimation via a Two-Stage Composite Likelihood | The multivariate probit is popular for modeling correlated binary data, with
an attractive balance of flexibility and simplicity. However, considerable
challenges remain in computation and in devising a clear statistical framework.
Interest in the multivariate probit has increased in recent years. Current
applications include genomics and precision medicine, where simultaneous
modeling of multiple traits may be of interest, and computational efficiency is
an important consideration. We propose a fast method for multivariate probit
estimation via a two-stage composite likelihood. We explore computational and
statistical efficiency, and note that the approach sets the stage for
extensions beyond the purely binary setting.
| stat |
Thresholded Graphical Lasso Adjusts for Latent Variables: Application to
Functional Neural Connectivity | In neuroscience, researchers seek to uncover the connectivity of neurons from
large-scale neural recordings or imaging; often people employ graphical model
selection and estimation techniques for this purpose. But, existing
technologies can only record from a small subset of neurons leading to a
challenging problem of graph selection in the presence of extensive latent
variables. Chandrasekaran et al. (2012) proposed a convex program to address
this problem that poses challenges from both a computational and statistical
perspective. To solve this problem, we propose an incredibly simple solution:
apply a hard thresholding operator to existing graph selection methods.
Conceptually simple and computationally attractive, we demonstrate that
thresholding the graphical Lasso, neighborhood selection, or CLIME estimators
have superior theoretical properties in terms of graph selection consistency as
well as stronger empirical results than existing approaches for the latent
variable graphical model problem. We also demonstrate the applicability of our
approach through a neuroscience case study on calcium-imaging data to estimate
functional neural connections.
| stat |
Computing the Expected Value of Sample Information Efficiently:
Expertise and Skills Required for Four Model-Based Methods | Objectives: Value of information (VOI) analyses can help policy-makers make
informed decisions about whether to conduct and how to design future studies.
Historically, a computationally expensive method to compute the Expected Value
of Sample Information (EVSI) restricted the use of VOI to simple decision
models and study designs. Recently, four EVSI approximation methods have made
such analyses more feasible and accessible. We provide practical
recommendations for analysts computing EVSI by evaluating these novel methods.
Methods: Members of the Collaborative Network for Value of Information (ConVOI)
compared the inputs, analyst's expertise and skills, and software required for
four recently developed approximation methods. Information was also collected
on the strengths and limitations of each approximation method. Results: All
four EVSI methods require a decision-analytic model's probabilistic sensitivity
analysis (PSA) output. One of the methods also requires the model to be re-run
to obtain new PSA outputs for each EVSI estimation. To compute EVSI, analysts
must be familiar with at least one of the following skills: advanced regression
modeling, likelihood specification, and Bayesian modeling. All methods have
different strengths and limitations, e.g., some methods handle evaluation of
study designs with more outcomes more efficiently while others quantify
uncertainty in EVSI estimates. All methods are programmed in the statistical
language R and two of the methods provide online applications. Conclusion: Our
paper helps to inform the choice between four efficient EVSI estimation
methods, enabling analysts to assess the methods' strengths and limitations and
select the most appropriate EVSI method given their situation and skills.
| stat |
Optimal low rank tensor recovery | We investigate the sample size requirement for exact recovery of a high order
tensor of low rank from a subset of its entries. In the Tucker decomposition
framework, we show that the Riemannian optimization algorithm with initial
value obtained from a spectral method can reconstruct a tensor of size $n\times
n \times\cdots \times n$ tensor of ranks $(r,\cdots,r)$ with high probability
from as few as $O((r^d+dnr)\log(d))$ entries. In the case of order 3 tensor,
the entries can be asymptotically as few as $O(nr)$ for a low rank large
tensor. We show the theoretical guarantee condition for the recovery. The
analysis relies on the tensor restricted isometry property (tensor RIP) and the
curvature of the low rank tensor manifold. Our algorithm is computationally
efficient and easy to implement. Numerical results verify that the algorithms
are able to recover a low rank tensor from minimum number of measurements. The
experiments on hyperspectral images recovery also show that our algorithm is
capable of real world signal processing problems.
| stat |
Boundary-free Estimators of the Mean Residual Life Function by
Transformation | We propose two new kernel-type estimators of the mean residual life function
$m_X(t)$ of bounded or half-bounded interval supported distributions. Though
not as severe as the boundary problems in the kernel density estimation,
eliminating the boundary bias problems that occur in the naive kernel estimator
of the mean residual life function is needed. In this article, we utilize the
property of bijective transformation. Furthermore, our proposed methods
preserve the mean value property, which cannot be done by the naive kernel
estimator. Some simulation results showing the estimators' performances and a
real data analysis will be presented in the last part of this article.
| stat |
Randomized Reduced Forward Models for Efficient Metropolis--Hastings
MCMC, with Application to Subsurface Fluid Flow and Capacitance Tomography | Bayesian modelling and computational inference by Markov chain Monte Carlo
(MCMC) is a principled framework for large-scale uncertainty quantification,
though is limited in practice by computational cost when implemented in the
simplest form that requires simulating an accurate computer model at each
iteration of the MCMC. The delayed acceptance Metropolis--Hastings MCMC
leverages a reduced model for the forward map to lower the compute cost per
iteration, though necessarily reduces statistical efficiency that can, without
care, lead to no reduction in the computational cost of computing estimates to
a desired accuracy. Randomizing the reduced model for the forward map can
dramatically improve computational efficiency, by maintaining the low cost per
iteration but also avoiding appreciable loss of statistical efficiency.
Randomized maps are constructed by a posteriori adaptive tuning of a randomized
and locally-corrected deterministic reduced model. Equivalently, the
approximated posterior distribution may be viewed as induced by a modified
likelihood function for use with the reduced map, with parameters tuned to
optimize the quality of the approximation to the correct posterior
distribution. Conditions for adaptive MCMC algorithms allow practical
approximations and algorithms that have guaranteed ergodicity for the target
distribution. Good statistical and computational efficiencies are demonstrated
in examples of calibration of large-scale numerical models of geothermal
reservoirs and electrical capacitance tomography.
| stat |
Particularities and commonalities of singular spectrum analysis as a
method of time series analysis and signal processing | Singular spectrum analysis (SSA), starting from the second half of the XX
century, has been a rapidly developing method of time series analysis. Since it
can be called principal component analysis for time series, SSA will definitely
be a standard method in time series analysis and signal processing in the
future. Moreover, the problems solved by SSA are considerably wider than that
for principal component analysis. In particular, the problems of frequency
estimation, forecasting and missing values imputation can be solved within the
framework of SSA. The idea of SSA came from different scientific communities,
such as that of researchers in time series analysis (Karhunen-Loeve
decomposition), signal processing (low-rank approximation and frequency
estimation) and multivariate data analysis (principal component analysis).
Also, depending on the area of applications, different viewpoints on the same
algorithms, choice of parameters, and methodology as a whole are considered.
Thus, the aim of the paper is to describe and compare different viewpoints on
SSA and its modifications and extensions to give people from different
scientific communities the possibility to be aware of potentially new aspects
of the method.
| stat |
Logistic Normal Multinomial Factor Analyzers for Clustering Microbiome
Data | The human microbiome plays an important role in human health and disease
status. Next generating sequencing technologies allow for quantifying the
composition of the human microbiome. Clustering these microbiome data can
provide valuable information by identifying underlying patterns across samples.
Recently, Fang and Subedi (2020) proposed a logistic normal multinomial mixture
model (LNM-MM) for clustering microbiome data. As microbiome data tends to be
high dimensional, here, we develop a family of logistic normal multinomial
factor analyzers (LNM-FA) by incorporating a factor analyzer structure in the
LNM-MM. This family of models is more suitable for high-dimensional data as the
number of parameters in LNM-FA can be greatly reduced by assuming that the
number of latent factors is small. Parameter estimation is done using a
computationally efficient variant of the alternating expectation conditional
maximization algorithm that utilizes variational Gaussian approximations. The
proposed method is illustrated using simulated and real datasets.
| stat |
Estimation of causal effects of multiple treatments in healthcare
database studies with rare outcomes | The preponderance of large-scale healthcare databases provide abundant
opportunities for comparative effectiveness research. Evidence necessary to
making informed treatment decisions often relies on comparing effectiveness of
multiple treatment options on outcomes of interest observed in a small number
of individuals. Causal inference with multiple treatments and rare outcomes is
a subject that has been treated sparingly in the literature. This paper designs
three sets of simulations, representative of the structure of our healthcare
database study, and propose causal analysis strategies for such settings. We
investigate and compare the operating characteristics of three types of methods
and their variants: Bayesian Additive Regression Trees (BART), regression
adjustment on multivariate spline of generalized propensity scores (RAMS) and
inverse probability of treatment weighting (IPTW) with multinomial logistic
regression or generalized boosted models. Our results suggest that BART and
RAMS provide lower bias and mean squared error, and the widely used IPTW
methods deliver unfavorable operating characteristics. We illustrate the
methods using a case study evaluating the comparative effectiveness of
robotic-assisted surgery, video-assisted thoracoscopic surgery and open
thoracotomy for treating non-small cell lung cancer.
| stat |
Randomization Tests to Assess Covariate Balance When Designing and
Analyzing Matched Datasets | Causal analyses for observational studies are often complicated by covariate
imbalances among treatment groups, and matching methodologies alleviate this
complication by finding subsets of treatment groups that exhibit covariate
balance. It is widely agreed upon that covariate balance can serve as evidence
that a matched dataset approximates a randomized experiment, but what kind of
experiment does a matched dataset approximate? In this work, we develop a
randomization test for the hypothesis that a matched dataset approximates a
particular experimental design, such as complete randomization, block
randomization, or rerandomization. Our test can incorporate any experimental
design, and it allows for a graphical display that puts several designs on the
same univariate scale, thereby allowing researchers to pinpoint which design --
if any -- is most appropriate for a matched dataset. After researchers
determine a plausible design, we recommend a randomization-based approach for
analyzing the matched data, which can incorporate any design and treatment
effect estimator. Through simulation, we find that our test can frequently
detect violations of randomized assignment that harm inferential results.
Furthermore, through simulation and a real application in political science, we
find that matched datasets with high levels of covariate balance tend to
approximate balance-constrained designs like rerandomization, and analyzing
them as such can lead to precise causal analyses. However, assuming a precise
design should be proceeded with caution, because it can harm inferential
results if there are still substantial biases due to remaining imbalances after
matching. Our approach is implemented in the randChecks R package, available on
CRAN.
| stat |
Indoor environment data time-series reconstruction using autoencoder
neural networks | As the number of installed meters in buildings increases, there is a growing
number of data time-series that could be used to develop data-driven models to
support and optimize building operation. However, building data sets are often
characterized by errors and missing values, which are considered, by the recent
research, among the main limiting factors on the performance of the proposed
models. Motivated by the need to address the problem of missing data in
building operation, this work presents a data-driven approach to fill these
gaps. In this study, three different autoencoder neural networks are trained to
reconstruct missing short-term indoor environment data time-series in a data
set collected in an office building in Aachen, Germany. This consisted of a
four year-long monitoring campaign in and between the years 2014 and 2017, of
84 different rooms. The models are applicable for different time-series
obtained from room automation, such as indoor air temperature, relative
humidity and $CO_{2}$ data streams. The results prove that the proposed methods
outperform classic numerical approaches and they result in reconstructing the
corresponding variables with average RMSEs of 0.42 {\deg}C, 1.30 % and 78.41
ppm, respectively.
| stat |
Multipopulation mortality modelling and forecasting: The multivariate
functional principal component with time weightings approaches | Human mortality patterns and trajectories in closely related populations are
likely linked together and share similarities. It is always desirable to model
them simultaneously while taking their heterogeneity into account. This paper
introduces two new models for joint mortality modelling and forecasting
multiple subpopulations in adaptations of the multivariate functional principal
component analysis techniques. The first model extends the independent
functional data model to a multi-population modelling setting. In the second
one, we propose a novel multivariate functional principal component method for
coherent modelling. Its design primarily fulfils the idea that when several
subpopulation groups have similar socio-economic conditions or common
biological characteristics, such close connections are expected to evolve in a
non-diverging fashion. We demonstrate the proposed methods by using
sex-specific mortality data. Their forecast performances are further compared
with several existing models, including the independent functional data model
and the Product-Ratio model, through comparisons with mortality data of ten
developed countries. Our experiment results show that the first proposed model
maintains a comparable forecast ability with the existing methods. In contrast,
the second proposed model outperforms the first model as well as the current
models in terms of forecast accuracy, in addition to several desirable
properties.
| stat |
PLS Generalized Linear Regression and Kernel Multilogit Algorithm (KMA)
for Microarray Data Classification | We implement extensions of the partial least squares generalized linear
regression (PLSGLR) due to Bastien et al. (2005) through its combination with
logistic regression and linear discriminant analysis, to get a partial least
squares generalized linear regression-logistic regression model (PLSGLR-log),
and a partial least squares generalized linear regression-linear discriminant
analysis model (PLSGLRDA). These two classification methods are then compared
with classical methodologies like the k-nearest neighbours (KNN), linear
discriminant analysis (LDA), partial least squares discriminant analysis
(PLSDA), ridge partial least squares (RPLS), and support vector machines(SVM).
Furthermore, we implement the kernel multilogit algorithm (KMA) by Dalmau et
al. (2015)and compare its performance with that of the other classifiers. The
results indicate that for both un-preprocessed and preprocessed data, the KMA
has the lowest classification error rates.
| stat |
Empirical Bayes Method for Boltzmann Machines | In this study, we consider an empirical Bayes method for Boltzmann machines
and propose an algorithm for it. The empirical Bayes method allows estimation
of the values of the hyperparameters of the Boltzmann machine by maximizing a
specific likelihood function referred to as the empirical Bayes likelihood
function in this study. However, the maximization is computationally hard
because the empirical Bayes likelihood function involves intractable
integrations of the partition function. The proposed algorithm avoids this
computational problem by using the replica method and the Plefka expansion. Our
method does not require any iterative procedures and is quite simple and fast,
though it introduces a bias to the estimate, which exhibits an unnatural
behavior with respect to the size of the dataset. This peculiar behavior is
supposed to be due to the approximate treatment by the Plefka expansion. A
possible extension to overcome this behavior is also discussed.
| stat |
Manifold valued data analysis of samples of networks, with applications
in corpus linguistics | Networks arise in many applications, such as in the analysis of text
documents, social interactions and brain activity. We develop a general
framework for extrinsic statistical analysis of samples of networks, motivated
by networks representing text documents in corpus linguistics. We identify
networks with their graph Laplacian matrices, for which we define metrics,
embeddings, tangent spaces, and a projection from Euclidean space to the space
of graph Laplacians. This framework provides a way of computing means,
performing principal component analysis and regression, and carrying out
hypothesis tests, such as for testing for equality of means between two samples
of networks. We apply the methodology to the set of novels by Jane Austen and
Charles Dickens.
| stat |
Dynamic Gene Coexpression Analysis with Correlation Modeling | In many transcriptomic studies, the correlation of genes might fluctuate with
quantitative factors such as genetic ancestry. We propose a method that models
the covariance between two variables to vary against a continuous covariate.
For the bivariate case, the proposed score test statistic is computationally
simple and robust to model misspecification of the covariance term.
Subsequently, the method is expanded to test relationships between one highly
connected gene, such as a transcription factor, and several other genes for a
more global investigation of the dynamic of the coexpression network.
Simulations show that the proposed method has higher statistical power than
alternatives, can be used in more diverse scenarios, and is computationally
cheaper. We apply this method to African American subjects from GTEx to analyze
the dynamic behavior of their gene coexpression against genetic ancestry and to
identify transcription factors whose coexpression with their target genes
change with the genetic ancestry. The proposed method can be applied to a wide
array of problems that require covariance modeling.
| stat |
Local Linear Forests | Random forests are a powerful method for non-parametric regression, but are
limited in their ability to fit smooth signals, and can show poor predictive
performance in the presence of strong, smooth effects. Taking the perspective
of random forests as an adaptive kernel method, we pair the forest kernel with
a local linear regression adjustment to better capture smoothness. The
resulting procedure, local linear forests, enables us to improve on asymptotic
rates of convergence for random forests with smooth signals, and provides
substantial gains in accuracy on both real and simulated data. We prove a
central limit theorem valid under regularity conditions on the forest and
smoothness constraints, and propose a computationally efficient construction
for confidence intervals. Moving to a causal inference application, we discuss
the merits of local regression adjustments for heterogeneous treatment effect
estimation, and give an example on a dataset exploring the effect word choice
has on attitudes to the social safety net. Last, we include simulation results
on real and generated data.
| stat |
General model-free weighted envelope estimation | Envelope methodology is succinctly pitched as a class of procedures for
increasing efficiency in multivariate analyses without altering traditional
objectives \citep[first sentence of page 1]{cook2018introduction}. This
description is true with the additional caveat that the efficiency gains
obtained by envelope methodology are mitigated by model selection volatility to
an unknown degree. The bulk of the current envelope methodology literature does
not account for this added variance that arises from the uncertainty in model
selection. Recent strides to account for model selection volatility have been
made on two fronts: 1) development of a weighted envelope estimator, in the
context of multivariate regression, to account for this variability directly;
2) development of a model selection criterion that facilitates consistent
estimation of the correct envelope model for more general settings. In this
paper, we unify these two directions and provide weighted envelope estimators
that directly account for the variability associated with model selection and
are appropriate for general multivariate estimation settings for vector valued
parameters. Our weighted estimation technique provides practitioners with
robust and useful variance reduction in finite samples. Theoretical
justification is given for our estimators and validity of a nonparametric
bootstrap procedure for estimating their asymptotic variance are established.
Simulation studies and a real data analysis support our claims and demonstrate
the advantage of our weighted envelope estimator when model selection
variability is present.
| stat |
Surrogate modeling of indoor down-link human exposure based on sparse
polynomial chaos expansion | Human exposure induced by wireless communication systems increasingly draws
the public attention. Here, an indoor down-link scenario is concerned and the
exposure level is statistically analyzed. The electromagnetic field (EMF)
emitted by a WiFi box is measured and electromagnetic dosimetry features are
evaluated from the whole-body specific absorption rate as computed with a
Finite-Difference Time-Domain (a.k.a. FDTD) code. Due to computational cost, a
statistical analysis is performed based on a surrogate model, which is
constructed by means of so-called sparse polynomial chaos expansion (PCE),
where the inner cross validation (ICV) is used to select the optimal
hyperparameters during the model construction and assess the model performance.
However, the ICV error is optimized and the model assessment tends to be overly
optimistic with small data sets. The method of cross-model validation is used
and outer cross validation is carried out for the model assessment. The effects
of the data preprocessing are investigated as well. Based on the surrogate
model, the global sensitivity of the exposure to input parameters is analyzed
from Sobol' indices.
| stat |
End-to-end Learning, with or without Labels | We present an approach for end-to-end learning that allows one to jointly
learn a feature representation from unlabeled data (with or without labeled
data) and predict labels for unlabeled data. The feature representation is
assumed to be specified in a differentiable programming framework, that is, as
a parameterized mapping amenable to automatic differentiation. The proposed
approach can be used with any amount of labeled and unlabeled data, gracefully
adjusting to the amount of supervision. We provide experimental results
illustrating the effectiveness of the approach.
| stat |
Target Robust Discriminant Analysis | In practice, the data distribution at test time often differs, to a smaller
or larger extent, from that of the original training data. Consequentially, the
so-called source classifier, trained on the available labelled data,
deteriorates on the test, or target, data. Domain adaptive classifiers aim to
combat this problem, but typically assume some particular form of domain shift.
Most are not robust to violations of domain shift assumptions and may even
perform worse than their non-adaptive counterparts. We construct robust
parameter estimators for discriminant analysis that guarantee performance
improvements of the adaptive classifier over the non-adaptive source
classifier.
| stat |
An Approximate Restricted Likelihood Ratio Test for Variance Components
in Generalized Linear Mixed Models | Generalized linear mixed models (GLMMs) are used to model responses from
exponential families with a combination of fixed and random effects. For
variance components in GLMMs, we propose an approximate restricted likelihood
ratio test that conducts testing on the working responses used in penalized
quasi-likelihood estimation. This presents the hypothesis test in terms of
normalized responses, allowing for application of existing testing methods for
linear mixed models. Our test is flexible, computationally efficient, and
outperforms several competitors. We illustrate the utility of the proposed
method with an extensive simulation study and two data applications. An R
package is provided.
| stat |
Approximately Optimal Spatial Design: How Good is it? | The increasing recognition of the association between adverse human health
conditions and many environmental substances as well as processes has led to
the need to monitor them. An important problem that arises in environmental
statistics is the design of the locations of the monitoring stations for those
environmental processes of interest. One particular design criterion for
monitoring networks that tries to reduce the uncertainty about predictions of
unseen processes is called the maximum-entropy design. However, this design
criterion involves a hard optimization problem that is computationally
intractable for large data sets. Previous work of Wang et al. (2017) examined a
probabilistic model that can be implemented efficiently to approximate the
underlying optimization problem. In this paper, we attempt to establish
statistically sound tools for assessing the quality of the approximations.
| stat |
Stacking with Neural network for Cryptocurrency investment | Predicting the direction of assets have been an active area of study and a
difficult task. Machine learning models have been used to build robust models
to model the above task. Ensemble methods is one of them showing results better
than a single supervised method. In this paper, we have used generative and
discriminative classifiers to create the stack, particularly 3 generative and 6
discriminative classifiers and optimized over one-layer Neural Network to model
the direction of price cryptocurrencies. Features used are technical indicators
used are not limited to trend, momentum, volume, volatility indicators, and
sentiment analysis has also been used to gain useful insight combined with the
above features. For Cross-validation, Purged Walk forward cross-validation has
been used. In terms of accuracy, we have done a comparative analysis of the
performance of Ensemble method with Stacking and Ensemble method with blending.
We have also developed a methodology for combined features importance for the
stacked model. Important indicators are also identified based on feature
importance.
| stat |
Penalized Interaction Estimation for Ultrahigh Dimensional Quadratic
Regression | Quadratic regression goes beyond the linear model by simultaneously including
main effects and interactions between the covariates. The problem of
interaction estimation in high dimensional quadratic regression has received
extensive attention in the past decade. In this article we introduce a novel
method which allows us to estimate the main effects and interactions
separately. Unlike existing methods for ultrahigh dimensional quadratic
regressions, our proposal does not require the widely used heredity assumption.
In addition, our proposed estimates have explicit formulas and obey the
invariance principle at the population level. We estimate the interactions of
matrix form under penalized convex loss function. The resulting estimates are
shown to be consistent even when the covariate dimension is an exponential
order of the sample size. We develop an efficient ADMM algorithm to implement
the penalized estimation. This ADMM algorithm fully explores the cheap
computational cost of matrix multiplication and is much more efficient than
existing penalized methods such as all pairs LASSO. We demonstrate the
promising performance of our proposal through extensive numerical studies.
| stat |
The Statistical Complexity of Early-Stopped Mirror Descent | Recently there has been a surge of interest in understanding implicit
regularization properties of iterative gradient-based optimization algorithms.
In this paper, we study the statistical guarantees on the excess risk achieved
by early-stopped unconstrained mirror descent algorithms applied to the
unregularized empirical risk with the squared loss for linear models and kernel
methods. By completing an inequality that characterizes convexity for the
squared loss, we identify an intrinsic link between offset Rademacher
complexities and potential-based convergence analysis of mirror descent
methods. Our observation immediately yields excess risk guarantees for the path
traced by the iterates of mirror descent in terms of offset complexities of
certain function classes depending only on the choice of the mirror map,
initialization point, step-size, and the number of iterations. We apply our
theory to recover, in a clean and elegant manner via rather short proofs, some
of the recent results in the implicit regularization literature, while also
showing how to improve upon them in some settings.
| stat |
Sequential Deconfounding for Causal Inference with Unobserved
Confounders | Using observational data to estimate the effect of a treatment is a powerful
tool for decision-making when randomized experiments are infeasible or costly.
However, observational data often yields biased estimates of treatment effects,
since treatment assignment can be confounded by unobserved variables. A remedy
is offered by deconfounding methods that adjust for such unobserved
confounders. In this paper, we develop the Sequential Deconfounder, a method
that enables estimating individualized treatment effects over time in presence
of unobserved confounders. This is the first deconfounding method that can be
used in a general sequential setting (i.e., with one or more treatments
assigned at each timestep). The Sequential Deconfounder uses a novel Gaussian
process latent variable model to infer substitutes for the unobserved
confounders, which are then used in conjunction with an outcome model to
estimate treatment effects over time. We prove that using our method yields
unbiased estimates of individualized treatment responses over time. Using
simulated and real medical data, we demonstrate the efficacy of our method in
deconfounding the estimation of treatment responses over time.
| stat |
Model-Robust Inference for Clinical Trials that Improve Precision by
Stratified Randomization and Covariate Adjustment | Two commonly used methods for improving precision and power in clinical
trials are stratified randomization and covariate adjustment. However, many
trials do not fully capitalize on the combined precision gains from these two
methods, which can lead to wasted resources in terms of sample size and trial
duration. We derive consistency and asymptotic normality of model-robust
estimators that combine these two methods, and show that these estimators can
lead to substantial gains in precision and power. Our theorems cover a class of
estimators that handle continuous, binary, and time-to-event outcomes; missing
outcomes under the missing at random assumption are handled as well. For each
estimator, we give a formula for a consistent variance estimator that is
model-robust and that fully captures variance reductions from stratified
randomization and covariate adjustment. Also, we give the first proof (to the
best of our knowledge) of consistency and asymptotic normality of the
Kaplan-Meier estimator under stratified randomization, and we derive its
asymptotic variance. The above results also hold for the biased-coin
covariate-adaptive design. We demonstrate our results using three completed,
phase 3, randomized trial data sets of treatments for substance use disorder,
where the variance reduction due to stratified randomization and covariate
adjustment ranges from 1% to 36%.
| stat |
A general perspective on the Metropolis-Hastings kernel | Since its inception the Metropolis-Hastings kernel has been applied in
sophisticated ways to address ever more challenging and diverse sampling
problems. Its success stems from the flexibility brought by the fact that its
verification and sampling implementation rests on a local ``detailed balance''
condition, as opposed to a global condition in the form of a typically
intractable integral equation. While checking the local condition is routine in
the simplest scenarios, this proves much more difficult for complicated
applications involving auxiliary structures and variables. Our aim is to
develop a framework making establishing correctness of complex Markov chain
Monte Carlo kernels a purely mechanical or algebraic exercise, while making
communication of ideas simpler and unambiguous by allowing a stronger focus on
essential features -- a choice of embedding distribution, an involution and
occasionally an acceptance function -- rather than the induced, boilerplate
structure of the kernels that often tends to obscure what is important. This
framework can also be used to validate kernels that do not satisfy detailed
balance, i.e. which are not reversible, but a modified version thereof.
| stat |
Mapping malaria seasonality: a case study from Madagascar | Many malaria-endemic areas experience seasonal fluctuations in case incidence
as Anopheles mosquito and Plasmodium parasite life cycles respond to changing
environmental conditions. While most existing maps of malaria seasonality use
fixed thresholds of rainfall, temperature, and/or vegetation indices to
identify suitable transmission months, we develop a statistical modelling
framework for characterising the seasonal patterns derived directly from case
data.
The procedure involves a spatiotemporal regression model for estimating the
monthly proportions of total annual cases and an algorithm to identify
operationally relevant characteristics such as the transmission start and peak
months. A seasonality index combines the monthly proportion estimates and
existing estimates of annual case incidence to provide a summary of "how
seasonal" locations are relative to their surroundings. An advancement upon
past seasonality mapping endeavours is the presentation of the uncertainty
associated with each map, which will enable policymakers to make more
statistically sound decisions. The methodology is illustrated using health
facility data from Madagascar.
| stat |
Individual dynamic prediction of clinical endpoint from large
dimensional longitudinal biomarker history: a landmark approach | The individual data collected throughout patient follow-up constitute crucial
information for assessing the risk of a clinical event, and eventually for
adapting a therapeutic strategy. Joint models and landmark models have been
proposed to compute individual dynamic predictions from repeated measures to
one or two markers. However, they hardly extend to the case where the complete
patient history includes much more repeated markers possibly. Our objective was
thus to propose a solution for the dynamic prediction of a health event that
may exploit repeated measures of a possibly large number of markers. We
combined a landmark approach extended to endogenous markers history with
machine learning methods adapted to survival data. Each marker trajectory is
modeled using the information collected up to landmark time, and summary
variables that best capture the individual trajectories are derived. These
summaries and additional covariates are then included in different prediction
methods. To handle a possibly large dimensional history, we rely on machine
learning methods adapted to survival data, namely regularized regressions and
random survival forests, to predict the event from the landmark time, and we
show how they can be combined into a superlearner. Then, the performances are
evaluated by cross-validation using estimators of Brier Score and the area
under the Receiver Operating Characteristic curve adapted to censored data. We
demonstrate in a simulation study the benefits of machine learning survival
methods over standard survival models, especially in the case of numerous
and/or nonlinear relationships between the predictors and the event. We then
applied the methodology in two prediction contexts: a clinical context with the
prediction of death for patients with primary biliary cholangitis, and a public
health context with the prediction of death in the general elderly population
at different ages. Our methodology, implemented in R, enables the prediction of
an event using the entire longitudinal patient history, even when the number of
repeated markers is large. Although introduced with mixed models for the
repeated markers and methods for a single right censored time-to-event, our
method can be used with any other appropriate modeling technique for the
markers and can be easily extended to competing risks setting.
| stat |
Metric Learning from Imbalanced Data | A key element of any machine learning algorithm is the use of a function that
measures the dis/similarity between data points. Given a task, such a function
can be optimized with a metric learning algorithm. Although this research field
has received a lot of attention during the past decade, very few approaches
have focused on learning a metric in an imbalanced scenario where the number of
positive examples is much smaller than the negatives. Here, we address this
challenging task by designing a new Mahalanobis metric learning algorithm (IML)
which deals with class imbalance. The empirical study performed shows the
efficiency of IML.
| stat |
Adaptive Clustering Using Kernel Density Estimators | We derive and analyze a generic, recursive algorithm for estimating all
splits in a finite cluster tree as well as the corresponding clusters. We
further investigate statistical properties of this generic clustering algorithm
when it receives level set estimates from a kernel density estimator. In
particular, we derive finite sample guarantees, consistency, rates of
convergence, and an adaptive data-driven strategy for choosing the kernel
bandwidth. For these results we do not need continuity assumptions on the
density such as H\"{o}lder continuity, but only require intuitive geometric
assumptions of non-parametric nature.
| stat |
Alternating Minimization Converges Super-Linearly for Mixed Linear
Regression | We address the problem of solving mixed random linear equations. We have
unlabeled observations coming from multiple linear regressions, and each
observation corresponds to exactly one of the regression models. The goal is to
learn the linear regressors from the observations. Classically, Alternating
Minimization (AM) (which is a variant of Expectation Maximization (EM)) is used
to solve this problem. AM iteratively alternates between the estimation of
labels and solving the regression problems with the estimated labels.
Empirically, it is observed that, for a large variety of non-convex problems
including mixed linear regression, AM converges at a much faster rate compared
to gradient based algorithms. However, the existing theory suggests similar
rate of convergence for AM and gradient based methods, failing to capture this
empirical behavior. In this paper, we close this gap between theory and
practice for the special case of a mixture of $2$ linear regressions. We show
that, provided initialized properly, AM enjoys a \emph{super-linear} rate of
convergence in certain parameter regimes. To the best of our knowledge, this is
the first work that theoretically establishes such rate for AM. Hence, if we
want to recover the unknown regressors upto an error (in $\ell_2$ norm) of
$\epsilon$, AM only takes $\mathcal{O}(\log \log (1/\epsilon))$ iterations.
Furthermore, we compare AM with a gradient based heuristic algorithm
empirically and show that AM dominates in iteration complexity as well as
wall-clock time.
| stat |
Spiked Laplacian Graphs: Bayesian Community Detection in Heterogeneous
Networks | In network data analysis, it is becoming common to work with a collection of
graphs that exhibit \emph{heterogeneity}. For example, neuroimaging data from
patient cohorts are increasingly available. A critical analytical task is to
identify communities, and graph Laplacian-based methods are routinely used.
However, these methods are currently limited to a single network and do not
provide measures of uncertainty on the community assignment. In this work, we
propose a probabilistic network model called the ``Spiked Laplacian Graph''
that considers each network as an invertible transform of the Laplacian, with
its eigenvalues modeled by a modified spiked structure. This effectively
reduces the number of parameters in the eigenvectors, and their sign patterns
allow efficient estimation of the community structure. Further, the posterior
distribution of the eigenvectors provides uncertainty quantification for the
community estimates. Subsequently, we introduce a Bayesian non-parametric
approach to address the issue of heterogeneity in a collection of graphs.
Theoretical results are established on the posterior consistency of the
procedure and provide insights on the trade-off between model resolution and
accuracy. We illustrate the performance of the methodology on synthetic data
sets, as well as a neuroscience study related to brain activity in working
memory.
Keywords: Hierarchical Community Detection, Isoperimetric Constant,
Mixed-Effect Eigendecomposition, Normalized Graph Cut, Stiefel Manifold
| stat |
UQ-CHI: An Uncertainty Quantification-Based Contemporaneous Health Index
for Degenerative Disease Monitoring | Developing knowledge-driven contemporaneous health index (CHI) that can
precisely reflect the underlying patient across the course of the condition's
progression holds a unique value, like facilitating a range of clinical
decision-making opportunities. This is particularly important for monitoring
degenerative condition such as Alzheimer's disease (AD), where the condition of
the patient will decay over time. Detecting early symptoms and progression
sign, and continuous severity evaluation, are all essential for disease
management. While a few methods have been developed in the literature,
uncertainty quantification of those health index models has been largely
neglected. To ensure the continuity of the care, we should be more explicit
about the level of confidence in model outputs. Ideally, decision-makers should
be provided with recommendations that are robust in the face of substantial
uncertainty about future outcomes. In this paper, we aim at filling this gap by
developing an uncertainty quantification based contemporaneous longitudinal
index, named UQ-CHI, with a particular focus on continuous patient monitoring
of degenerative conditions. Our method is to combine convex optimization and
Bayesian learning using the maximum entropy learning (MEL) framework,
integrating uncertainty on labels as well. Our methodology also provides
closed-form solutions in some important decision making tasks, e.g., such as
predicting the label of a new sample. Numerical studies demonstrate the
effectiveness of the propose UQ-CHI method in prediction accuracy, monitoring
efficacy, and unique advantages if uncertainty quantification is enabled
practice.
| stat |
Semiparametric Bayesian Inference for Local Extrema of Functions in the
Presence of Noise | There is a wide range of applications where the local extrema of a function
are the key quantity of interest. However, there is surprisingly little work on
methods to infer local extrema with uncertainty quantification in the presence
of noise. By viewing the function as an infinite-dimensional nuisance
parameter, a semiparametric formulation of this problem poses daunting
challenges, both methodologically and theoretically, as (i) the number of local
extrema may be unknown, and (ii) the induced shape constraints associated with
local extrema are highly irregular. In this article, we address these
challenges by suggesting an encompassing strategy that eliminates the need to
specify the number of local extrema, which leads to a remarkably simple, fast
semiparametric Bayesian approach for inference on local extrema. We provide
closed-form characterization of the posterior distribution and study its large
sample behaviors under this encompassing regime. We show a multi-modal
Bernstein-von Mises phenomenon in which the posterior measure converges to a
mixture of Gaussians with the number of components matching the underlying
truth, leading to posterior exploration that accounts for multi-modality. We
illustrate the method through simulations and a real data application to
event-related potential analysis.
| stat |
On the Inductive Bias of Neural Tangent Kernels | State-of-the-art neural networks are heavily over-parameterized, making the
optimization algorithm a crucial ingredient for learning predictive models with
good generalization properties. A recent line of work has shown that in a
certain over-parameterized regime, the learning dynamics of gradient descent
are governed by a certain kernel obtained at initialization, called the neural
tangent kernel. We study the inductive bias of learning in such a regime by
analyzing this kernel and the corresponding function space (RKHS). In
particular, we study smoothness, approximation, and stability properties of
functions with finite norm, including stability to image deformations in the
case of convolutional networks, and compare to other known kernels for similar
architectures.
| stat |
Bayesian imaging using Plug & Play priors: when Langevin meets Tweedie | Since the seminal work of Venkatakrishnan et al. (2013), Plug & Play (PnP)
methods have become ubiquitous in Bayesian imaging. These methods derive
Minimum Mean Square Error (MMSE) or Maximum A Posteriori (MAP) estimators for
inverse problems in imaging by combining an explicit likelihood function with a
prior that is implicitly defined by an image denoising algorithm. The PnP
algorithms proposed in the literature mainly differ in the iterative schemes
they use for optimisation or for sampling. In the case of optimisation schemes,
some recent works guarantee the convergence to a fixed point, albeit not
necessarily a MAP estimate. In the case of sampling schemes, to the best of our
knowledge, there is no known proof of convergence. There also remain important
open questions regarding whether the underlying Bayesian models and estimators
are well defined, well-posed, and have the basic regularity properties required
to support these numerical schemes. To address these limitations, this paper
develops theory, methods, and provably convergent algorithms for performing
Bayesian inference with PnP priors. We introduce two algorithms: 1) PnP-ULA
(Unadjusted Langevin Algorithm) for Monte Carlo sampling and MMSE inference;
and 2) PnP-SGD (Stochastic Gradient Descent) for MAP inference. Using recent
results on the quantitative convergence of Markov chains, we establish detailed
convergence guarantees for these two algorithms under realistic assumptions on
the denoising operators used, with special attention to denoisers based on deep
neural networks. We also show that these algorithms approximately target a
decision-theoretically optimal Bayesian model that is well-posed. The proposed
algorithms are demonstrated on several canonical problems such as image
deblurring, inpainting, and denoising, where they are used for point estimation
as well as for uncertainty visualisation and quantification.
| stat |
Deep Involutive Generative Models for Neural MCMC | We introduce deep involutive generative models, a new architecture for deep
generative modeling, and use them to define Involutive Neural MCMC, a new
approach to fast neural MCMC. An involutive generative model represents a
probability kernel $G(\phi \mapsto \phi')$ as an involutive (i.e.,
self-inverting) deterministic function $f(\phi, \pi)$ on an enlarged state
space containing auxiliary variables $\pi$. We show how to make these models
volume preserving, and how to use deep volume-preserving involutive generative
models to make valid Metropolis-Hastings updates based on an auxiliary variable
scheme with an easy-to-calculate acceptance ratio. We prove that deep
involutive generative models and their volume-preserving special case are
universal approximators for probability kernels. This result implies that with
enough network capacity and training time, they can be used to learn
arbitrarily complex MCMC updates. We define a loss function and optimization
algorithm for training parameters given simulated data. We also provide initial
experiments showing that Involutive Neural MCMC can efficiently explore
multi-modal distributions that are intractable for Hybrid Monte Carlo, and can
converge faster than A-NICE-MC, a recently introduced neural MCMC technique.
| stat |
Quantification of the weight of fingerprint evidence using a ROC-based
Approximate Bayesian Computation algorithm for model selection | For more than a century, fingerprints have been used with considerable
success to identify criminals or verify the identity of individuals. The
categorical conclusion scheme used by fingerprint examiners, and more generally
the inference process followed by forensic scientists, have been heavily
criticised in the scientific and legal literature. Instead, scholars have
proposed to characterise the weight of forensic evidence using the Bayes factor
as the key element of the inference process. In forensic science, quantifying
the magnitude of support is equally as important as determining which model is
supported. Unfortunately, the complexity of fingerprint patterns render
likelihood-based inference impossible. In this paper, we use an Approximate
Bayesian Computation model selection algorithm to quantify the weight of
fingerprint evidence. We supplement the ABC algorithm using a Receiver
Operating Characteristic curve to mitigate the effect of the curse of
dimensionality. Our modified algorithm is computationally efficient and makes
it easier to monitor convergence as the number of simulations increase. We use
our method to quantify the weight of fingerprint evidence in forensic science,
but we note that it can be applied to any other forensic pattern evidence.
| stat |
Sketching Datasets for Large-Scale Learning (long version) | This article considers "compressive learning," an approach to large-scale
machine learning where datasets are massively compressed before learning (e.g.,
clustering, classification, or regression) is performed. In particular, a
"sketch" is first constructed by computing carefully chosen nonlinear random
features (e.g., random Fourier features) and averaging them over the whole
dataset. Parameters are then learned from the sketch, without access to the
original dataset. This article surveys the current state-of-the-art in sketched
learning, including the main concepts and algorithms, their connections with
established signal-processing methods, existing theoretical guarantees -- on
both information preservation and privacy preservation, and important open
problems.
| stat |
A Review of Generalizability and Transportability | When assessing causal effects, determining the target population to which the
results are intended to generalize is a critical decision. Randomized and
observational studies each have strengths and limitations for estimating causal
effects in a target population. Estimates from randomized data may have
internal validity but are often not representative of the target population.
Observational data may better reflect the target population, and hence be more
likely to have external validity, but are subject to potential bias due to
unmeasured confounding. While much of the causal inference literature has
focused on addressing internal validity bias, both internal and external
validity are necessary for unbiased estimates in a target population. This
paper presents a framework for addressing external validity bias, including a
synthesis of approaches for generalizability and transportability, the
assumptions they require, as well as tests for the heterogeneity of treatment
effects and differences between study and target populations.
| stat |
The Shape of Data: Intrinsic Distance for Data Distributions | The ability to represent and compare machine learning models is crucial in
order to quantify subtle model changes, evaluate generative models, and gather
insights on neural network architectures. Existing techniques for comparing
data distributions focus on global data properties such as mean and covariance;
in that sense, they are extrinsic and uni-scale. We develop a first-of-its-kind
intrinsic and multi-scale method for characterizing and comparing data
manifolds, using a lower-bound of the spectral variant of the
Gromov-Wasserstein inter-manifold distance, which compares all data moments. In
a thorough experimental study, we demonstrate that our method effectively
discerns the structure of data manifolds even on unaligned data of different
dimensionalities; moreover, we showcase its efficacy in evaluating the quality
of generative models.
| stat |
Stochastic Gradient Annealed Importance Sampling for Efficient Online
Marginal Likelihood Estimation | We consider estimating the marginal likelihood in settings with independent
and identically distributed (i.i.d.) data. We propose estimating the predictive
distributions in a sequential factorization of the marginal likelihood in such
settings by using stochastic gradient Markov Chain Monte Carlo techniques. This
approach is far more efficient than traditional marginal likelihood estimation
techniques such as nested sampling and annealed importance sampling due to its
use of mini-batches to approximate the likelihood. Stability of the estimates
is provided by an adaptive annealing schedule. The resulting stochastic
gradient annealed importance sampling (SGAIS) technique, which is the key
contribution of our paper, enables us to estimate the marginal likelihood of a
number of models considerably faster than traditional approaches, with no
noticeable loss of accuracy. An important benefit of our approach is that the
marginal likelihood is calculated in an online fashion as data becomes
available, allowing the estimates to be used for applications such as online
weighted model combination.
| stat |
SSNdesign -- an R package for pseudo-Bayesian optimal and adaptive
sampling designs on stream networks | Streams and rivers are biodiverse and provide valuable ecosystem services.
Maintaining these ecosystems is an important task, so organisations often
monitor the status and trends in stream condition and biodiversity using field
sampling and, more recently, autonomous in-situ sensors. However, data
collection is often costly and so effective and efficient survey designs are
crucial to maximise information while minimising costs. Geostatistics and
optimal and adaptive design theory can be used to optimise the placement of
sampling sites in freshwater studies and aquatic monitoring programs.
Geostatistical modelling and experimental design on stream networks pose
statistical challenges due to the branching structure of the network, flow
connectivity and directionality, and differences in flow volume. Thus, unique
challenges of geostatistics and experimental design on stream networks
necessitates the development of new open-source software for implementing the
theory. We present SSNdesign, an R package for solving optimal and adaptive
design problems on stream networks that integrates with existing open-source
software. We demonstrate the mathematical foundations of our approach, and
illustrate the functionality of SSNdesign using two case studies involving real
data from Queensland, Australia. In both case studies we demonstrate that the
optimal or adaptive designs outperform random and spatially balanced survey
designs. The SSNdesign package has the potential to boost the efficiency of
freshwater monitoring efforts and provide much-needed information for
freshwater conservation and management.
| stat |
What to make of non-inferiority and equivalence testing with a
post-specified margin? | In order to determine whether or not an effect is absent based on a
statistical test, the recommended frequentist tool is the equivalence test.
Typically, it is expected that an appropriate equivalence margin has been
specified before any data are observed. Unfortunately, this can be a difficult
task. If the margin is too small, then the test's power will be substantially
reduced. If the margin is too large, any claims of equivalence will be
meaningless. Moreover, it remains unclear how defining the margin afterwards
will bias one's results. In this short article, we consider a series of
hypothetical scenarios in which the margin is defined post-hoc or is otherwise
considered controversial. We also review a number of relevant, potentially
problematic actual studies from clinical trials research, with the aim of
motivating a critical discussion as to what is acceptable and desirable in the
reporting and interpretation of equivalence tests.
| stat |
A Comparison of Model-Free Phase I Dose Escalation Designs for
Dual-Agent Combination Therapies | It is increasingly common for therapies in oncology to be given in
combination. In some cases, patients can benefit from the interaction between
two drugs, although often at the risk of higher toxicity. A large number of
designs to conduct phase I trials in this setting are available, where the
objective is to select the maximum tolerated dose combination (MTC). Recently,
a number of model-free (also called model-assisted) designs have provoked
interest, providing several practical advantages over the more conventional
approaches of rule-based or model-based designs. In this paper, we demonstrate
a novel calibration procedure for model-free designs to determine their most
desirable parameters. Under the calibration procedure, we compare the behaviour
of model-free designs to a model-based approach in a comprehensive simulation
study, covering a number of clinically plausible scenarios. It is found that
model-free designs are competitive with the model-based design in terms of the
proportion of correct selections of the MTC. However, there are a number of
scenarios in which model-free designs offer a safer alternative. This is also
illustrated in the application of the designs to a case study using data from a
phase I oncology trial.
| stat |
A general theory for preferential sampling in environmental networks | This paper presents a general model framework for detecting the preferential
sampling of environmental monitors recording an environmental process across
space and/or time. This is achieved by considering the joint distribution of an
environmental process with a site--selection process that considers where and
when sites are placed to measure the process. The environmental process may be
spatial, temporal or spatio--temporal in nature. By sharing random effects
between the two processes, the joint model is able to establish whether site
placement was stochastically dependent of the environmental process under
study. The embedding into a spatio--temporal framework also allows for the
modelling of the dynamic site---selection process itself. Real--world factors
affecting both the size and location of the network can be easily modelled and
quantified. Depending upon the choice of population of locations to consider
for selection across space and time under the site--selection process,
different insights about the precise nature of preferential sampling can be
obtained. The general framework developed in the paper is designed to be easily
and quickly fit using the R-INLA package. We apply this framework to a case
study involving particulate air pollution over the UK where a major reduction
in the size of a monitoring network through time occurred. It is demonstrated
that a significant response--biased reduction in the air quality monitoring
network occurred. We also show that the network was consistently
unrepresentative of the levels of particulate matter seen across much of GB
throughout the operating life of the network. Finally we show that this may
have led to a severe over-reporting of the population--average exposure levels
experienced across GB. This could have great impacts on estimates of the health
effects of black smoke levels.
| stat |
Estimating Number of Factors by Adjusted Eigenvalues Thresholding | Determining the number of common factors is an important and practical topic
in high dimensional factor models. The existing literatures are mainly based on
the eigenvalues of the covariance matrix. Due to the incomparability of the
eigenvalues of the covariance matrix caused by heterogeneous scales of observed
variables, it is very difficult to give an accurate relationship between these
eigenvalues and the number of common factors. To overcome this limitation, we
appeal to the correlation matrix and show surprisingly that the number of
eigenvalues greater than $1$ of population correlation matrix is the same as
the number of common factors under some mild conditions. To utilize such a
relationship, we study the random matrix theory based on the sample correlation
matrix in order to correct the biases in estimating the top eigenvalues and to
take into account of estimation errors in eigenvalue estimation. This leads us
to propose adjusted correlation thresholding (ACT) for determining the number
of common factors in high dimensional factor models, taking into account the
sampling variabilities and biases of top sample eigenvalues. We also establish
the optimality of the proposed methods in terms of minimal signal strength and
optimal threshold. Simulation studies lend further support to our proposed
method and show that our estimator outperforms other competing methods in most
of our testing cases.
| stat |
A simulation comparison of tournament designs for the World Men's
Handball Championships | The study aims to compare different designs for the World Men's Handball
Championships. This event, organised in every two years, has adopted four
hybrid formats consisting of knockout and round-robin stages in recent decades,
including a change of design between the two recent championships in 2017 and
2019. They are evaluated under two extremal seeding policies with respect to
various outcome measures through Monte-Carlo simulations. We find that the
ability to give the first four positions to the strongest teams, as well as the
expected quality and outcome uncertainty of the final is not necessarily a
monotonic function of the number of matches played: the most frugal format is
the second best with respect to these outcome measures, making it a good
compromise in an unavoidable trade-off. A possible error is identified in a
particular design. The relative performance of the formats is independent of
the seeding rules and the competitive balance of the teams. The recent reform
is demonstrated to have increased the probability of winning for the top teams.
Our results have useful implications for the organisers of hybrid tournaments.
| stat |
Robust Approximate Bayesian Inference with Synthetic Likelihood | Bayesian synthetic likelihood (BSL) is now an established method for
conducting approximate Bayesian inference in models where, due to the
intractability of the likelihood function, exact Bayesian approaches are either
infeasible or computationally too demanding. Implicit in the application of BSL
is the assumption that the data generating process (DGP) can produce simulated
summary statistics that capture the behaviour of the observed summary
statistics. We demonstrate that if this compatibility between the actual and
assumed DGP is not satisfied, i.e., if the model is misspecified, BSL can yield
unreliable parameter inference. To circumvent this issue, we propose a new BSL
approach that can detect the presence of model misspecification, and
simultaneously deliver useful inferences even under significant model
misspecification. Two simulated and two real data examples demonstrate the
performance of this new approach to BSL, and document its superior accuracy
over standard BSL when the assumed model is misspecified.
| stat |
Adding experimental treatment arms to Multi-Arm Multi-Stage platform
trials in progress | Multi-Arm Multi-Stage (MAMS) platform trials are an efficient tool for the
comparison of several treatments with a control. Suppose a new treatment
becomes available at some stage of a trial already in progress. There are clear
benefits to adding the treatment to the current trial for comparison, but how?
As flexible as the MAMS framework is, it requires pre-planned options for how
the trial proceeds at each stage in order to control the familywise error rate.
Thus, as with many adaptive designs, it is difficult to make unplanned design
modifications. The conditional error approach is a tool that allows unplanned
design modifications while maintaining the overall error rate. In this work we
use the conditional error approach to allow adding new arms to a MAMS trial in
progress.
Using a single stage two-arm trial, we demonstrate the principals of
incorporating additional hypotheses into the testing structure. With this
framework for adding treatments and hypotheses in place, we show how to update
the testing procedure for a MAMS trial in progress to incorporate additional
treatment arms. Through simulation, we illustrate the operating characteristics
of such procedures.
| stat |
Analyzing and Improving Representations with the Soft Nearest Neighbor
Loss | We explore and expand the $\textit{Soft Nearest Neighbor Loss}$ to measure
the $\textit{entanglement}$ of class manifolds in representation space: i.e.,
how close pairs of points from the same class are relative to pairs of points
from different classes. We demonstrate several use cases of the loss. As an
analytical tool, it provides insights into the evolution of class similarity
structures during learning. Surprisingly, we find that $\textit{maximizing}$
the entanglement of representations of different classes in the hidden layers
is beneficial for discrimination in the final layer, possibly because it
encourages representations to identify class-independent similarity structures.
Maximizing the soft nearest neighbor loss in the hidden layers leads not only
to improved generalization but also to better-calibrated estimates of
uncertainty on outlier data. Data that is not from the training distribution
can be recognized by observing that in the hidden layers, it has fewer than the
normal number of neighbors from the predicted class.
| stat |
Stochastic modified equations for the asynchronous stochastic gradient
descent | We propose a stochastic modified equations (SME) for modeling the
asynchronous stochastic gradient descent (ASGD) algorithms. The resulting SME
of Langevin type extracts more information about the ASGD dynamics and
elucidates the relationship between different types of stochastic gradient
algorithms. We show the convergence of ASGD to the SME in the continuous time
limit, as well as the SME's precise prediction to the trajectories of ASGD with
various forcing terms. As an application of the SME, we propose an optimal
mini-batching strategy for ASGD via solving the optimal control problem of the
associated SME.
| stat |
Causal Meta-Mediation Analysis: Inferring Dose-Response Function From
Summary Statistics of Many Randomized Experiments | It is common in the internet industry to use offline-developed algorithms to
power online products that contribute to the success of a business.
Offline-developed algorithms are guided by offline evaluation metrics, which
are often different from online business key performance indicators (KPIs). To
maximize business KPIs, it is important to pick a north star among all
available offline evaluation metrics. By noting that online products can be
measured by online evaluation metrics, the online counterparts of offline
evaluation metrics, we decompose the problem into two parts. As the offline A/B
test literature works out the first part: counterfactual estimators of offline
evaluation metrics that move the same way as their online counterparts, we
focus on the second part: causal effects of online evaluation metrics on
business KPIs. The north star of offline evaluation metrics should be the one
whose online counterpart causes the most significant lift in the business KPI.
We model the online evaluation metric as a mediator and formalize its causality
with the business KPI as dose-response function (DRF). Our novel approach,
causal meta-mediation analysis, leverages summary statistics of many existing
randomized experiments to identify, estimate, and test the mediator DRF. It is
easy to implement and to scale up, and has many advantages over the literature
of mediation analysis and meta-analysis. We demonstrate its effectiveness by
simulation and implementation on real data.
| stat |
Statistical Estimation of Ergodic Markov Chain Kernel over Discrete
State Space | We investigate the statistical complexity of estimating the parameters of a
discrete-state Markov chain kernel from a single long sequence of state
observations. In the finite case, we characterize (modulo logarithmic factors)
the minimax sample complexity of estimation with respect to the operator
infinity norm, while in the countably infinite case, we analyze the problem
with respect to a natural entry-wise norm derived from total variation. We show
that in both cases, the sample complexity is governed by the mixing properties
of the unknown chain, for which, in the finite-state case, there are known
finite-sample estimators with fully empirical confidence intervals.
| stat |
CONQ: CONtinuous Quantile Treatment Effects for Large-Scale Online
Controlled Experiments | In many industry settings, online controlled experimentation (A/B test) has
been broadly adopted as the gold standard to measure product or feature
impacts. Most research has primarily focused on user engagement type metrics,
specifically measuring treatment effects at mean (average treatment effects,
ATE), and only a few have been focusing on performance metrics (e.g. latency),
where treatment effects are measured at quantiles. Measuring quantile treatment
effects (QTE) is challenging due to the myriad difficulties such as dependency
introduced by clustered samples, scalability issues, density bandwidth choices,
etc. In addition, previous literature has mainly focused on QTE at some
pre-defined locations, such as P50 or P90, which doesn't always convey the full
picture. In this paper, we propose a novel scalable non-parametric solution,
which can provide a continuous range of QTE with point-wise confidence
intervals while circumventing the density estimation altogether. Numerical
results show high consistency with traditional methods utilizing asymptotic
normality. An end-to-end pipeline has been implemented at Snap Inc., providing
daily insights on key performance metrics at a distributional level.
| stat |
Cumulative deviation of a subpopulation from the full population | Assessing equity in treatment of a subpopulation often involves assigning
numerical "scores" to all individuals in the full population such that similar
individuals get similar scores; matching via propensity scores or appropriate
covariates is common, for example. Given such scores, individuals with similar
scores may or may not attain similar outcomes independent of the individuals'
memberships in the subpopulation. The traditional graphical methods for
visualizing inequities are known as "reliability diagrams" or "calibrations
plots," which bin the scores into a partition of all possible values, and for
each bin plot both the average outcomes for only individuals in the
subpopulation as well as the average outcomes for all individuals; comparing
the graph for the subpopulation with that for the full population gives some
sense of how the averages for the subpopulation deviate from the averages for
the full population. Unfortunately, real data sets contain only finitely many
observations, limiting the usable resolution of the bins, and so the
conventional methods can obscure important variations due to the binning.
Fortunately, plotting cumulative deviation of the subpopulation from the full
population as proposed in this paper sidesteps the problematic coarse binning.
The cumulative plots encode subpopulation deviation directly as the slopes of
secant lines for the graphs. Slope is easy to perceive even when the constant
offsets of the secant lines are irrelevant. The cumulative approach avoids
binning that smooths over deviations of the subpopulation from the full
population. Such cumulative aggregation furnishes both high-resolution
graphical methods and simple scalar summary statistics (analogous to those of
Kuiper and of Kolmogorov and Smirnov used in statistical significance testing
for comparing probability distributions).
| stat |
A robust approach to model-based classification based on trimming and
constraints | In a standard classification framework a set of trustworthy learning data are
employed to build a decision rule, with the final aim of classifying unlabelled
units belonging to the test set. Therefore, unreliable labelled observations,
namely outliers and data with incorrect labels, can strongly undermine the
classifier performance, especially if the training size is small. The present
work introduces a robust modification to the Model-Based Classification
framework, employing impartial trimming and constraints on the ratio between
the maximum and the minimum eigenvalue of the group scatter matrices. The
proposed method effectively handles noise presence in both response and
exploratory variables, providing reliable classification even when dealing with
contaminated datasets. A robust information criterion is proposed for model
selection. Experiments on real and simulated data, artificially adulterated,
are provided to underline the benefits of the proposed method.
| stat |
Private Tabular Survey Data Products through Synthetic Microdata
Generation | We propose three synthetic microdata approaches to generate private tabular
survey data products for public release. We adapt a disclosure risk
based-weighted pseudo posterior mechanism to survey data with a focus on
producing tabular products under a formal privacy guarantee. Two of our
approaches synthesize the observed sample distribution of the outcome and
survey weights, jointly, such that both quantities together possess a
probabilistic differential privacy guarantee. The privacy-protected outcome and
sampling weights are used to construct tabular cell estimates and associated
standard errors to correct for survey sampling bias. The third approach
synthesizes the population distribution from the observed sample under a pseudo
posterior construction that treats survey sampling weights as fixed to correct
the sample likelihood to approximate that for the population. Each by-record
sampling weight in the pseudo posterior is, in turn, multiplied by the
associated privacy, risk-based weight for that record to create a composite
pseudo posterior mechanism that both corrects for survey bias and provides a
privacy guarantee for the observed sample. Through a simulation study and a
real data application to the Survey of Doctorate Recipients public use file, we
demonstrate that our three microdata synthesis approaches to construct tabular
products provide superior utility preservation as compared to the
additive-noise approach of the Laplace Mechanism. Moreover, all our approaches
allow the release of microdata to the public, enabling additional analyses at
no extra privacy cost.
| stat |
ROC and AUC with a Binary Predictor: a Potentially Misleading Metric | In analysis of binary outcomes, the receiver operator characteristic (ROC)
curve is heavily used to show the performance of a model or algorithm. The ROC
curve is informative about the performance over a series of thresholds and can
be summarized by the area under the curve (AUC), a single number. When a
predictor is categorical, the ROC curve has one less than number of categories
as potential thresholds; when the predictor is binary there is only one
threshold. As the AUC may be used in decision-making processes on determining
the best model, it important to discuss how it agrees with the intuition from
the ROC curve. We discuss how the interpolation of the curve between thresholds
with binary predictors can largely change the AUC. Overall, we show using a
linear interpolation from the ROC curve with binary predictors corresponds to
the estimated AUC, which is most commonly done in software, which we believe
can lead to misleading results. We compare R, Python, Stata, and SAS software
implementations. We recommend using reporting the interpolation used and
discuss the merit of using the step function interpolator, also referred to as
the "pessimistic" approach by Fawcett (2006).
| stat |
Scalable Population Synthesis with Deep Generative Modeling | Population synthesis is concerned with the generation of synthetic yet
realistic representations of populations. It is a fundamental problem in the
modeling of transport where the synthetic populations of micro-agents represent
a key input to most agent-based models. In this paper, a new methodological
framework for how to 'grow' pools of micro-agents is presented. The model
framework adopts a deep generative modeling approach from machine learning
based on a Variational Autoencoder (VAE). Compared to the previous population
synthesis approaches, including Iterative Proportional Fitting (IPF), Gibbs
sampling and traditional generative models such as Bayesian Networks or Hidden
Markov Models, the proposed method allows fitting the full joint distribution
for high dimensions. The proposed methodology is compared with a conventional
Gibbs sampler and a Bayesian Network by using a large-scale Danish trip diary.
It is shown that, while these two methods outperform the VAE in the
low-dimensional case, they both suffer from scalability issues when the number
of modeled attributes increases. It is also shown that the Gibbs sampler
essentially replicates the agents from the original sample when the required
conditional distributions are estimated as frequency tables. In contrast, the
VAE allows addressing the problem of sampling zeros by generating agents that
are virtually different from those in the original data but have similar
statistical properties. The presented approach can support agent-based modeling
at all levels by enabling richer synthetic populations with smaller zones and
more detailed individual characteristics.
| stat |
swdpwr: A SAS Macro and An R Package for Power Calculation in Stepped
Wedge Cluster Randomized Trials | Conditional and marginal models can both be utilized in power calculations
for stepped wedge cluster randomized trials. Hussey and Hughes (2007) first
provided the design and analysis for this particular type of cluster randomized
trials based on the linear mixed effects model for continuous outcomes, along
with an approximation to this approach for binary outcomes. Although not
accurate for binary outcomes, it has been widely used. To improve the
approximation for binary outcomes, Zhou, Liao, Kunz, Normand, Wang, and
Spiegelman (2020) and Li, Turner, and Preisser (2018b) have recently published
two new methods for stepped wedge designs (SWDs) of binary outcomes. However,
these new methods have not been implemented in publicly available software. In
this work, we have developed a SAS macro %swdpwr and an R package swdpwr to
calculate power in SWDs in various settings for both continuous and binary
outcomes.
| stat |
Extending the step-size restriction for gradient descent to avoid strict
saddle points | We provide larger step-size restrictions for which gradient descent based
algorithms (almost surely) avoid strict saddle points. In particular, consider
a twice differentiable (non-convex) objective function whose gradient has
Lipschitz constant L and whose Hessian is well-behaved. We prove that the
probability of initial conditions for gradient descent with step-size up to 2/L
converging to a strict saddle point, given one uniformly random initialization,
is zero. This extends previous results up to the sharp limit imposed by the
convex case. In addition, the arguments hold in the case when a learning rate
schedule is given, with either a continuous decaying rate or a piece-wise
constant schedule.
| stat |
Bayesian Estimation of Mixed Multinomial Logit Models: Advances and
Simulation-Based Evaluations | Variational Bayes (VB) methods have emerged as a fast and
computationally-efficient alternative to Markov chain Monte Carlo (MCMC)
methods for scalable Bayesian estimation of mixed multinomial logit (MMNL)
models. It has been established that VB is substantially faster than MCMC at
practically no compromises in predictive accuracy. In this paper, we address
two critical gaps concerning the usage and understanding of VB for MMNL. First,
extant VB methods are limited to utility specifications involving only
individual-specific taste parameters. Second, the finite-sample properties of
VB estimators and the relative performance of VB, MCMC and maximum simulated
likelihood estimation (MSLE) are not known. To address the former, this study
extends several VB methods for MMNL to admit utility specifications including
both fixed and random utility parameters. To address the latter, we conduct an
extensive simulation-based evaluation to benchmark the extended VB methods
against MCMC and MSLE in terms of estimation times, parameter recovery and
predictive accuracy. The results suggest that all VB variants with the
exception of the ones relying on an alternative variational lower bound
constructed with the help of the modified Jensen's inequality perform as well
as MCMC and MSLE at prediction and parameter recovery. In particular, VB with
nonconjugate variational message passing and the delta-method (VB-NCVMP-Delta)
is up to 16 times faster than MCMC and MSLE. Thus, VB-NCVMP-Delta can be an
attractive alternative to MCMC and MSLE for fast, scalable and accurate
estimation of MMNL models.
| stat |
On the Complexity of Approximating Multimarginal Optimal Transport | We study the complexity of approximating the multimarginal optimal transport
(MOT) distance, a generalization of the classical optimal transport distance,
considered here between $m$ discrete probability distributions supported each
on $n$ support points. First, we show that the standard linear programming (LP)
representation of the MOT problem is not a minimum-cost flow problem when $m
\geq 3$. This negative result implies that some combinatorial algorithms, e.g.,
network simplex method, are not suitable for approximating the MOT problem,
while the worst-case complexity bound for the deterministic interior-point
algorithm remains a quantity of $\tilde{O}(n^{3m})$. We then propose two simple
and \textit{deterministic} algorithms for approximating the MOT problem. The
first algorithm, which we refer to as \textit{multimarginal Sinkhorn}
algorithm, is a provably efficient multimarginal generalization of the Sinkhorn
algorithm. We show that it achieves a complexity bound of
$\tilde{O}(m^3n^m\varepsilon^{-2})$ for a tolerance $\varepsilon \in (0, 1)$.
This provides a first \textit{near-linear time} complexity bound guarantee for
approximating the MOT problem and matches the best known complexity bound for
the Sinkhorn algorithm in the classical OT setting when $m = 2$. The second
algorithm, which we refer to as \textit{accelerated multimarginal Sinkhorn}
algorithm, achieves the acceleration by incorporating an estimate sequence and
the complexity bound is $\tilde{O}(m^3n^{m+1/3}\varepsilon^{-4/3})$. This bound
is better than that of the first algorithm in terms of $1/\varepsilon$, and
accelerated alternating minimization
algorithm~\citep{Tupitsa-2020-Multimarginal} in terms of $n$. Finally, we
compare our new algorithms with the commercial LP solver \textsc{Gurobi}.
Preliminary results on synthetic data and real images demonstrate the
effectiveness and efficiency of our algorithms.
| stat |
Power Analysis for Stepped Wedge Trials with Two Treatments | Stepped wedge designs (SWDs) are designs for cluster randomized trials that
feature staggered, unidirectional cross-over, typically from a control to a
treatment condition. Existing literature on statistical power for SWDs
primarily focuses on designs with a single treatment. However, SWDs with
multiple treatments are being proposed and conducted. We present a linear mixed
model for a SWD with two treatments, with and without an interaction between
them. We derive closed form solutions for the standard errors of the treatment
effect coefficients for such models along with power calculation methods. We
consider repeated cross-sectional designs as well as open and closed cohort
designs and different random effect structures. Design features are examined to
determine their impact on power for main treatment and interaction effects.
| stat |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.