title
stringlengths 6
244
| abstract
stringlengths 19
5.09k
| label
stringclasses 10
values |
|---|---|---|
A Simplified Stochastic EM Algorithm for Negative Binomial Cure Rate
Model and Comparison with EM Algorithm | In this paper, a long-term survival model under competing risks is
considered. The unobserved number of competing risks is assumed to follow a
negative binomial distribution that can capture both over- and
under-dispersion. Considering the latent competing risks as missing data, a
variation of the well-known expectation maximization (EM) algorithm, called the
stochastic EM algorithm (SEM), is developed. It is shown that the SEM algorithm
avoids calculation of complicated expectations, which is a major advantage of
the SEM algorithm over the EM algorithm. The proposed procedure also allows the
objective function to be split into two simpler functions, one corresponding to
the parameters associated with the cure rate and the other corresponding to the
parameters associated with the progression times. The advantage of this
approach is that each simple function, with lower parameter dimension, can be
maximized independently. An extensive Monte Carlo simulation study is carried
out to compare the performances of the SEM and EM algorithms. Finally, a breast
cancer survival data is analyzed and it is shown that the SEM algorithm
performs better than the EM algorithm.
| stat |
Solar Radiation Anomaly Events Modeling Using Spatial-Temporal Mutually
Interactive Processes | Modeling and predicting solar events, in particular, the solar ramping event
is critical for improving situational awareness for solar power generation
systems. Solar ramping events are significantly impacted by weather conditions
such as temperature, humidity, and cloud density. Discovering the correlation
between different locations and times is a highly challenging task since the
system is complex and noisy. We propose a novel method to model and predict
ramping events from spatial-temporal sequential solar radiation data based on a
spatio-temporal interactive Bernoulli process. We demonstrate the good
performance of our approach on real solar radiation datasets.
| stat |
Directed Graphical Models and Causal Discovery for Zero-Inflated Data | Modern RNA sequencing technologies provide gene expression measurements from
single cells that promise refined insights on regulatory relationships among
genes. Directed graphical models are well-suited to explore such (cause-effect)
relationships. However, statistical analyses of single cell data are
complicated by the fact that the data often show zero-inflated expression
patterns. To address this challenge, we propose directed graphical models that
are based on Hurdle conditional distributions parametrized in terms of
polynomials in parent variables and their 0/1 indicators of being zero or
nonzero. While directed graphs for Gaussian models are only identifiable up to
an equivalence class in general, we show that, under a natural and weak
assumption, the exact directed acyclic graph of our zero-inflated models can be
identified. We propose methods for graph recovery, apply our model to real
single-cell RNA-seq data on T helper cells, and show simulated experiments that
validate the identifiability and graph estimation methods in practice.
| stat |
Noisy and Incomplete Boolean Matrix Factorizationvia Expectation
Maximization | Probabilistic approach to Boolean matrix factorization can provide solutions
robustagainst noise and missing values with linear computational complexity.
However,the assumption about latent factors can be problematic in real world
applications.This study proposed a new probabilistic algorithm free of
assumptions of latentfactors, while retaining the advantages of previous
algorithms. Real data experimentshowed that our algorithm was favourably
compared with current state-of-the-artprobabilistic algorithms.
| stat |
Automatic Hyperparameter Tuning Method for Local Outlier Factor, with
Applications to Anomaly Detection | In recent years, there have been many practical applications of anomaly
detection such as in predictive maintenance, detection of credit fraud, network
intrusion, and system failure. The goal of anomaly detection is to identify in
the test data anomalous behaviors that are either rare or unseen in the
training data. This is a common goal in predictive maintenance, which aims to
forecast the imminent faults of an appliance given abundant samples of normal
behaviors. Local outlier factor (LOF) is one of the state-of-the-art models
used for anomaly detection, but the predictive performance of LOF depends
greatly on the selection of hyperparameters. In this paper, we propose a novel,
heuristic methodology to tune the hyperparameters in LOF. A tuned LOF model
that uses the proposed method shows good predictive performance in both
simulations and real data sets.
| stat |
Tree-Projected Gradient Descent for Estimating Gradient-Sparse
Parameters on Graphs | We study estimation of a gradient-sparse parameter vector
$\boldsymbol{\theta}^* \in \mathbb{R}^p$, having strong gradient-sparsity
$s^*:=\|\nabla_G \boldsymbol{\theta}^*\|_0$ on an underlying graph $G$. Given
observations $Z_1,\ldots,Z_n$ and a smooth, convex loss function $\mathcal{L}$
for which $\boldsymbol{\theta}^*$ minimizes the population risk
$\mathbb{E}[\mathcal{L}(\boldsymbol{\theta};Z_1,\ldots,Z_n)]$, we propose to
estimate $\boldsymbol{\theta}^*$ by a projected gradient descent algorithm that
iteratively and approximately projects gradient steps onto spaces of vectors
having small gradient-sparsity over low-degree spanning trees of $G$. We show
that, under suitable restricted strong convexity and smoothness assumptions for
the loss, the resulting estimator achieves the squared-error risk
$\frac{s^*}{n} \log (1+\frac{p}{s^*})$ up to a multiplicative constant that is
independent of $G$. In contrast, previous polynomial-time algorithms have only
been shown to achieve this guarantee in more specialized settings, or under
additional assumptions for $G$ and/or the sparsity pattern of $\nabla_G
\boldsymbol{\theta}^*$. As applications of our general framework, we apply our
results to the examples of linear models and generalized linear models with
random design.
| stat |
Generalised Bayesian Structural Equation Modelling | We propose a generalised framework for Bayesian Structural Equation Modelling
(SEM) that can be applied to a variety of data types. The introduced framework
focuses on the approximate zero approach, according to which a hypothesised
structure is formulated with approximate rather than exact zero. It extends
previously suggested models by \citeA{MA12} and can handle continuous, binary,
and ordinal data. Moreover, we propose a novel model assessment paradigm aiming
to address shortcomings of posterior predictive $p-$values, which provide the
default metric of fit for Bayesian SEM. The introduced model assessment
procedure monitors the out-of-sample predictive performance of the model in
question, and draws from a list of principles to answer whether the
hypothesised theory is supported by the data. We incorporate scoring rules and
cross-validation to supplement existing model assessment metrics for Bayesian
SEM. The methodology is illustrated in continuous and categorical data examples
via simulation experiments as well as real-world applications on the `Big-5'
personality scale and the Fagerstrom test for nicotine dependence.
| stat |
The probability of a robust inference for internal validity and its
applications in regression models | The internal validity of observational study is often subject to debate. In
this study, we define the unobserved sample based on the counterfactuals and
formalize its relationship with the null hypothesis statistical testing (NHST)
for regression models. The probability of a robust inference for internal
validity, i.e., the PIV, is the probability of rejecting the null hypothesis
again based on the ideal sample which is defined as the combination of the
observed and unobserved samples, provided the same null hypothesis has already
been rejected for the observed sample. When the unconfoundedness assumption is
dubious, one can bound the PIV of an inference based on bounded belief about
the mean counterfactual outcomes, which is often needed in this case.
Essentially, the PIV is statistical power of the NHST that is thought to be
built on the ideal sample. We summarize the process of evaluating internal
validity with the PIV into a six-step procedure and illustrate it with an
empirical example (i.e., Hong and Raudenbush (2005)).
| stat |
Seemingly Unrelated Regression with Measurement Error: Estimation via
Markov chain Monte Carlo and Mean Field Variational Bayes Approximation | Linear regression with measurement error in the covariates is a heavily
studied topic, however, the statistics/econometrics literature is almost silent
to estimating a multi-equation model with measurement error. This paper
considers a seemingly unrelated regression model with measurement error in the
covariates and introduces two novel estimation methods: a pure Bayesian
algorithm (based on Markov chain Monte Carlo techniques) and its mean field
variational Bayes (MFVB) approximation. The MFVB method has the added advantage
of being computationally fast and can handle big data. An issue pertinent to
measurement error models is parameter identification, and this is resolved by
employing a prior distribution on the measurement error variance. The methods
are shown to perform well in multiple simulation studies, where we analyze the
impact on posterior estimates arising due to different values of reliability
ratio or variance of the true unobserved quantity used in the data generating
process. The paper further implements the proposed algorithms in an application
drawn from the health literature and shows that modeling measurement error in
the data can improve model fitting.
| stat |
Randomization tests for peer effects in group formation experiments | Measuring the effect of peers on individual outcomes is a challenging
problem, in part because individuals often select peers who are similar in both
observable and unobservable ways. Group formation experiments avoid this
problem by randomly assigning individuals to groups and observing their
responses; for example, do first-year students have better grades when they are
randomly assigned roommates who have stronger academic backgrounds? Standard
approaches for analyzing these experiments, however, are heavily
model-dependent and generally fail to exploit the randomized design. In this
paper, we extend methods from randomization-based testing under interference to
group formation experiments. The proposed tests are justified by the
randomization itself, require relatively few assumptions, and are exact in
finite samples. First, we develop procedures that yield valid tests for
arbitrary group formation designs. Second, we derive sufficient conditions on
the design such that the randomization test can be implemented via simple
random permutations. We apply this approach to two recent group formation
experiments and implement the proposed method in the new RGroupFormation R
package.
| stat |
Adaptive surrogate models for parametric studies | The computational effort for the evaluation of numerical simulations based on
e.g. the finite-element method is high. Metamodels can be utilized to create a
low-cost alternative. However the number of required samples for the creation
of a sufficient metamodel should be kept low, which can be achieved by using
adaptive sampling techniques. In this Master thesis adaptive sampling
techniques are investigated for their use in creating metamodels with the
Kriging technique, which interpolates values by a Gaussian process governed by
prior covariances. The Kriging framework with extension to multifidelity
problems is presented and utilized to compare adaptive sampling techniques
found in the literature for benchmark problems as well as applications for
contact mechanics. This thesis offers the first comprehensive comparison of a
large spectrum of adaptive techniques for the Kriging framework. Furthermore a
multitude of adaptive techniques is introduced to multifidelity Kriging as well
as well as to a Kriging model with reduced hyperparameter dimension called
partial least squares Kriging. In addition, an innovative adaptive scheme for
binary classification is presented and tested for identifying chaotic motion of
a Duffing's type oscillator.
| stat |
Estimate Sequences for Stochastic Composite Optimization: Variance
Reduction, Acceleration, and Robustness to Noise | In this paper, we propose a unified view of gradient-based algorithms for
stochastic convex composite optimization by extending the concept of estimate
sequence introduced by Nesterov. More precisely, we interpret a large class of
stochastic optimization methods as procedures that iteratively minimize a
surrogate of the objective, which covers the stochastic gradient descent method
and variants of the incremental approaches SAGA, SVRG, and MISO/Finito/SDCA.
This point of view has several advantages: (i) we provide a simple generic
proof of convergence for all of the aforementioned methods; (ii) we naturally
obtain new algorithms with the same guarantees; (iii) we derive generic
strategies to make these algorithms robust to stochastic noise, which is useful
when data is corrupted by small random perturbations. Finally, we propose a new
accelerated stochastic gradient descent algorithm and an accelerated SVRG
algorithm with optimal complexity that is robust to stochastic noise.
| stat |
Evaluating epidemic forecasts in an interval format | For practical reasons, many forecasts of case, hospitalization and death
counts in the context of the current COVID-19 pandemic are issued in the form
of central predictive intervals at various levels. This is also the case for
the forecasts collected in the COVID-19 Forecast Hub
(https://covid19forecasthub.org/). Forecast evaluation metrics like the
logarithmic score, which has been applied in several infectious disease
forecasting challenges, are then not available as they require full predictive
distributions. This article provides an overview of how established methods for
the evaluation of quantile and interval forecasts can be applied to epidemic
forecasts in this format. Specifically, we discuss the computation and
interpretation of the weighted interval score, which is a proper score that
approximates the continuous ranked probability score. It can be interpreted as
a generalization of the absolute error to probabilistic forecasts and allows
for a decomposition into a measure of sharpness and penalties for over- and
underprediction.
| stat |
Information Geometry of Orthogonal Initializations and Training | Recently mean field theory has been successfully used to analyze properties
of wide, random neural networks. It gave rise to a prescriptive theory for
initializing feed-forward neural networks with orthogonal weights, which
ensures that both the forward propagated activations and the backpropagated
gradients are near $\ell_2$ isometries and as a consequence training is orders
of magnitude faster. Despite strong empirical performance, the mechanisms by
which critical initializations confer an advantage in the optimization of deep
neural networks are poorly understood. Here we show a novel connection between
the maximum curvature of the optimization landscape (gradient smoothness) as
measured by the Fisher information matrix (FIM) and the spectral radius of the
input-output Jacobian, which partially explains why more isometric networks can
train much faster. Furthermore, given that orthogonal weights are necessary to
ensure that gradient norms are approximately preserved at initialization, we
experimentally investigate the benefits of maintaining orthogonality throughout
training, from which we conclude that manifold optimization of weights performs
well regardless of the smoothness of the gradients. Moreover, motivated by
experimental results we show that a low condition number of the FIM is not
predictive of faster learning.
| stat |
Fast Maximum Likelihood Estimation and Supervised Classification for the
Beta-Liouville Multinomial | The multinomial and related distributions have long been used to model
categorical, count-based data in fields ranging from bioinformatics to natural
language processing. Commonly utilized variants include the standard
multinomial and the Dirichlet multinomial distributions due to their
computational efficiency and straightforward parameter estimation process.
However, these distributions make strict assumptions about the mean, variance,
and covariance between the categorical features being modeled. If these
assumptions are not met by the data, it may result in poor parameter estimates
and loss in accuracy for downstream applications like classification. Here, we
explore efficient parameter estimation and supervised classification methods
using an alternative distribution, called the Beta-Liouville multinomial, which
relaxes some of the multinomial assumptions. We show that the Beta-Liouville
multinomial is comparable in efficiency to the Dirichlet multinomial for
Newton-Raphson maximum likelihood estimation, and that its performance on
simulated data matches or exceeds that of the multinomial and Dirichlet
multinomial distributions. Finally, we demonstrate that the Beta-Liouville
multinomial outperforms the multinomial and Dirichlet multinomial on two out of
four gold standard datasets, supporting its use in modeling data with low to
medium class overlap in a supervised classification context.
| stat |
BUDD: Multi-modal Bayesian Updating Deforestation Detections | The global phenomenon of forest degradation is a pressing issue with severe
implications for climate stability and biodiversity protection. In this work we
generate Bayesian updating deforestation detection (BUDD) algorithms by
incorporating Sentinel-1 backscatter and interferometric coherence with
Sentinel-2 normalized vegetation index data. We show that the algorithm
provides good performance in validation AOIs. We compare the effectiveness of
different combinations of the three data modalities as inputs into the BUDD
algorithm and compare against existing benchmarks based on optical imagery.
| stat |
Scalable Panel Fusion Using Distributed Min Cost Flow | Modern audience measurement requires combining observations from disparate
panel datasets. Connecting and relating such panel datasets is a process termed
panel fusion. This paper formalizes the panel fusion problem and presents a
novel approach to solve it. We cast the panel fusion as a network flow problem,
allowing the application of a rich body of research. In the context of digital
audience measurement, where panel sizes can grow into the tens of millions, we
propose an efficient algorithm to partition the network into sub-problems.
While the algorithm solves a relaxed version of the original problem, we
provide conditions under which it guarantees optimality. We demonstrate our
approach by fusing two real-world panel datasets in a distributed computing
environment.
| stat |
Nonparametric Density Estimation from Markov Chains | We introduce a new nonparametric density estimator inspired by Markov Chains,
and generalizing the well-known Kernel Density Estimator (KDE). Our estimator
presents several benefits with respect to the usual ones and can be used
straightforwardly as a foundation in all density-based algorithms. We prove the
consistency of our estimator and we find it typically outperforms KDE in
situations of large sample size and high dimensionality. We also employ our
density estimator to build a local outlier detector, showing very promising
results when applied to some realistic datasets.
| stat |
Statistical Inference for Polyak-Ruppert Averaged Zeroth-order
Stochastic Gradient Algorithm | As machine learning models are deployed in critical applications, it becomes
important to not just provide point estimators of the model parameters (or
subsequent predictions), but also quantify the uncertainty associated with
estimating the model parameters via confidence sets. In the last decade,
estimating or training in several machine learning models has become synonymous
with running stochastic gradient algorithms. However, computing the stochastic
gradients in several settings is highly expensive or even impossible at times.
An important question which has thus far not been addressed sufficiently in the
statistical machine learning literature is that of equipping zeroth-order
stochastic gradient algorithms with practical yet rigorous inferential
capabilities. Towards this, in this work, we first establish a central limit
theorem for Polyak-Ruppert averaged stochastic gradient algorithm in the
zeroth-order setting. We then provide online estimators of the asymptotic
covariance matrix appearing in the central limit theorem, thereby providing a
practical procedure for constructing asymptotically valid confidence sets (or
intervals) for parameter estimation (or prediction) in the zeroth-order
setting.
| stat |
Spatially dependent mixture models via the Logistic Multivariate CAR
prior | We consider the problem of spatially dependent areal data, where for each
area independent observations are available, and propose to model the density
of each area through a finite mixture of Gaussian distributions. The spatial
dependence is introduced via a novel joint distribution for a collection of
vectors in the simplex, that we term logisticMCAR. We show that salient
features of the logisticMCAR distribution can be described analytically, and
that a suitable augmentation scheme based on the P\'olya-Gamma identity allows
to derive an efficient Markov Chain Monte Carlo algorithm. When compared to
competitors, our model has proved to better estimate densities in different
(disconnected) areal locations when they have different characteristics. We
discuss an application on a real dataset of Airbnb listings in the city of
Amsterdam, also showing how to easily incorporate for additional covariate
information in the model.
| stat |
Adversarial Optimal Transport Through The Convolution Of Kernels With
Evolving Measures | A novel algorithm is proposed to solve the sample-based optimal transport
problem. An adversarial formulation of the push-forward condition uses a test
function built as a convolution between an adaptive kernel and an evolving
probability distribution $\nu$ over a latent variable $b$. Approximating this
convolution by its simulation over evolving samples $b^i(t)$ of $\nu$, the
parameterization of the test function reduces to determining the flow of these
samples. This flow, discretized over discrete time steps $t_n$, is built from
the composition of elementary maps. The optimal transport also follows a flow
that, by duality, must follow the gradient of the test function. The
representation of the test function as the Monte Carlo simulation of a
distribution makes the algorithm robust to dimensionality, and its evolution
under a memory-less flow produces rich, complex maps from simple parametric
transformations. The algorithm is illustrated with numerical examples.
| stat |
High-dimensional Gaussian sampling: a review and a unifying approach
based on a stochastic proximal point algorithm | Efficient sampling from a high-dimensional Gaussian distribution is an old
but high-stake issue. Vanilla Cholesky samplers imply a computational cost and
memory requirements which can rapidly become prohibitive in high dimension. To
tackle these issues, multiple methods have been proposed from different
communities ranging from iterative numerical linear algebra to Markov chain
Monte Carlo (MCMC) approaches. Surprisingly, no complete review and comparison
of these methods have been conducted. This paper aims at reviewing all these
approaches by pointing out their differences, close relations, benefits and
limitations. In addition to this state of the art, this paper proposes a
unifying Gaussian simulation framework by deriving a stochastic counterpart of
the celebrated proximal point algorithm in optimization. This framework offers
a novel and unifying revisit of most of the existing MCMC approaches while
extending them. Guidelines to choose the appropriate Gaussian simulation method
for a given sampling problem in high dimension are proposed and illustrated
with numerical examples.
| stat |
Testing and Support Recovery of Correlation Structures for Matrix-Valued
Observations with an Application to Stock Market Data | Estimation of the covariance matrix of asset returns is crucial to portfolio
construction. As suggested by economic theories, the correlation structure
among assets differs between emerging markets and developed countries. It is
therefore imperative to make rigorous statistical inference on correlation
matrix equality between the two groups of countries. However, if the
traditional vector-valued approach is undertaken, such inference is either
infeasible due to limited number of countries comparing to the relatively
abundant assets, or invalid due to the violations of temporal independence
assumption. This highlights the necessity of treating the observations as
matrix-valued rather than vector-valued. With matrix-valued observations, our
problem of interest can be formulated as statistical inference on covariance
structures under matrix normal distributions, i.e., testing independence and
correlation equality, as well as the corresponding support estimations. We
develop procedures that are asymptotically optimal under some regularity
conditions. Simulation results demonstrate the computational and statistical
advantages of our procedures over certain existing state-of-the-art methods.
Application of our procedures to stock market data validates several economic
propositions.
| stat |
Deep Importance Sampling based on Regression for Model Inversion and
Emulation | Understanding systems by forward and inverse modeling is a recurrent topic of
research in many domains of science and engineering. In this context, Monte
Carlo methods have been widely used as powerful tools for numerical inference
and optimization. They require the choice of a suitable proposal density that
is crucial for their performance. For this reason, several adaptive importance
sampling (AIS) schemes have been proposed in the literature. We here present an
AIS framework called Regression-based Adaptive Deep Importance Sampling
(RADIS). In RADIS, the key idea is the adaptive construction via regression of
a non-parametric proposal density (i.e., an emulator), which mimics the
posterior distribution and hence minimizes the mismatch between proposal and
target densities. RADIS is based on a deep architecture of two (or more) nested
IS schemes, in order to draw samples from the constructed emulator. The
algorithm is highly efficient since employs the posterior approximation as
proposal density, which can be improved adding more support points. As a
consequence, RADIS asymptotically converges to an exact sampler under mild
conditions. Additionally, the emulator produced by RADIS can be in turn used as
a cheap surrogate model for further studies. We introduce two specific RADIS
implementations that use Gaussian Processes (GPs) and Nearest Neighbors (NN)
for constructing the emulator. Several numerical experiments and comparisons
show the benefits of the proposed schemes. A real-world application in remote
sensing model inversion and emulation confirms the validity of the approach.
| stat |
Robust subspace recovery by Tyler's M-estimator | This paper considers the problem of robust subspace recovery: given a set of
$N$ points in $\mathbb{R}^D$, if many lie in a $d$-dimensional subspace, then
can we recover the underlying subspace? We show that Tyler's M-estimator can be
used to recover the underlying subspace, if the percentage of the inliers is
larger than $d/D$ and the data points lie in general position. Empirically,
Tyler's M-estimator compares favorably with other convex subspace recovery
algorithms in both simulations and experiments on real data sets.
| stat |
Improvement of Batch Normalization in Imbalanced Data | In this study, we consider classification problems based on neural networks
in data-imbalanced environment. Learning from an imbalanced data set is one of
the most important and practical problems in the field of machine learning. A
weighted loss function based on cost-sensitive approach is a well-known
effective method for imbalanced data sets. We consider a combination of
weighted loss function and batch normalization (BN) in this study. BN is a
powerful standard technique in the recent developments in deep learning. A
simple combination of both methods leads to a size-mismatch problem due to a
mismatch between interpretations of effective size of data set in both methods.
We propose a simple modification to BN to correct the size-mismatch and
demonstrate that this modified BN is effective in data-imbalanced environment.
| stat |
Frequentism-as-model | Most statisticians are aware that probability models interpreted in a
frequentist manner are not really true in objective reality, but only
idealisations. I argue that this is often ignored when actually applying
frequentist methods and interpreting the results, and that keeping up the
awareness for the essential difference between reality and models can lead to a
more appropriate use and interpretation of frequentist models and methods,
called frequentism-as-model. This is elaborated showing connections to existing
work, appreciating the special role of i.i.d. models and subject matter
knowledge, giving an account of how and under what conditions models that are
not true can be useful, giving detailed interpretations of tests and confidence
intervals, confronting their implicit compatibility logic with the inverse
probability logic of Bayesian inference, re-interpreting the role of model
assumptions, appreciating robustness, and the role of ``interpretative
equivalence'' of models. Epistemic (often referred to as Bayesian) probability
shares the issue that its models are only idealisations and not really true for
modelling reasoning about uncertainty, meaning that it does not have an
essential advantage over frequentism, as is often claimed. Bayesian statistics
can be combined with frequentism-as-model, leading to what Gelman and Hennig
(2017) call ``falsificationist Bayes''.
| stat |
Doubly Stochastic Generative Arrivals Modeling | We propose a new framework named DS-WGAN that integrates the doubly
stochastic (DS) structure and the Wasserstein generative adversarial networks
(WGAN) to model, estimate, and simulate a wide class of arrival processes with
general non-stationary and random arrival rates. Regarding statistical
properties, we prove consistency and convergence rate for the estimator solved
by the DS-WGAN framework under a non-parametric smoothness condition. Regarding
computational efficiency and tractability, we address a challenge in gradient
evaluation and model estimation, arised from the discontinuity in the
simulator. We then show that the DS-WGAN framework can conveniently facilitate
what-if simulation and predictive simulation for future scenarios that are
different from the history. Numerical experiments with synthetic and real data
sets are implemented to demonstrate the performance of DS-WGAN. The performance
is measured from both a statistical perspective and an operational performance
evaluation perspective. Numerical experiments suggest that, in terms of
performance, the successful model estimation for DS-WGAN only requires a
moderate size of representative data, which can be appealing in many contexts
of operational management.
| stat |
Bayesian Hierarchical Models for the Prediction of Volleyball Results | Statistical modelling of sports data has become more and more popular in the
recent years and different types of models have been proposed to achieve a
variety of objectives: from identifying the key characteristics which lead a
team to win or lose to predicting the outcome of a game or the team rankings in
national leagues. Although not as popular as football or basketball, volleyball
is a team sport with both national and international level competitions in
almost every country. However, there is almost no study investigating the
prediction of volleyball game outcomes and team rankings in national leagues.
We propose a Bayesian hierarchical model for the prediction of the rankings of
volleyball national teams, which also allows to estimate the results of each
match in the league. We consider two alternative model specifications of
different complexity which are validated using data from the women's volleyball
Italian Serie A1 2017-2018 season.
| stat |
Causal Inference in Geoscience and Remote Sensing from Observational
Data | Establishing causal relations between random variables from observational
data is perhaps the most important challenge in today's \blue{science}. In
remote sensing and geosciences this is of special relevance to better
understand the Earth's system and the complex interactions between the
governing processes. In this paper, we focus on observational causal inference,
thus we try to estimate the correct direction of causation using a finite set
of empirical data. In addition, we focus on the more complex bivariate scenario
that requires strong assumptions and no conditional independence tests can be
used. In particular, we explore the framework of (non-deterministic) additive
noise models, which relies on the principle of independence between the cause
and the generating mechanism. A practical algorithmic instantiation of such
principle only requires 1) two regression models in the forward and backward
directions, and 2) the estimation of {\em statistical independence} between the
obtained residuals and the observations. The direction leading to more
independent residuals is decided to be the cause. We instead propose a
criterion that uses the {\em sensitivity} (derivative) of the dependence
estimator, the sensitivity criterion allows to identify samples most affecting
the dependence measure, and hence the criterion is robust to spurious
detections. We illustrate performance in a collection of 28 geoscience causal
inference problems, in a database of radiative transfer models simulations and
machine learning emulators in vegetation parameter modeling involving 182
problems, and in assessing the impact of different regression models in a
carbon cycle problem. The criterion achieves state-of-the-art detection rates
in all cases, it is generally robust to noise sources and distortions.
| stat |
Classifying Treatment Responders Under Causal Effect Monotonicity | In the context of individual-level causal inference, we study the problem of
predicting whether someone will respond or not to a treatment based on their
features and past examples of features, treatment indicator (e.g., drug/no
drug), and a binary outcome (e.g., recovery from disease). As a classification
task, the problem is made difficult by not knowing the example outcomes under
the opposite treatment indicators. We assume the effect is monotonic, as in
advertising's effect on a purchase or bail-setting's effect on reappearance in
court: either it would have happened regardless of treatment, not happened
regardless, or happened only depending on exposure to treatment. Predicting
whether the latter is latently the case is our focus. While previous work
focuses on conditional average treatment effect estimation, formulating the
problem as a classification task rather than an estimation task allows us to
develop new tools more suited to this problem. By leveraging monotonicity, we
develop new discriminative and generative algorithms for the
responder-classification problem. We explore and discuss connections to
corrupted data and policy learning. We provide an empirical study with both
synthetic and real datasets to compare these specialized algorithms to standard
benchmarks.
| stat |
An algorithm for non-parametric estimation in state-space models | State-space models are ubiquitous in the statistical literature since they
provide a flexible and interpretable framework for analyzing many time series.
In most practical applications, the state-space model is specified through a
parametric model. However, the specification of such a parametric model may
require an important modeling effort or may lead to models which are not
flexible enough to reproduce all the complexity of the phenomenon of interest.
In such situations, an appealing alternative consists in inferring the
state-space model directly from the data using a non-parametric framework. The
recent developments of powerful simulation techniques have permitted to improve
the statistical inference for parametric state-space models. It is proposed to
combine two of these techniques, namely the Stochastic Expectation-Maximization
(SEM) algorithm and Sequential Monte Carlo (SMC) approaches, for non-parametric
estimation in state-space models. The performance of the proposed algorithm is
assessed through simulations on toy models and an application to environmental
data is discussed.
| stat |
Validating Bayesian Inference Algorithms with Simulation-Based
Calibration | Verifying the correctness of Bayesian computation is challenging. This is
especially true for complex models that are common in practice, as these
require sophisticated model implementations and algorithms. In this paper we
introduce \emph{simulation-based calibration} (SBC), a general procedure for
validating inferences from Bayesian algorithms capable of generating posterior
samples. This procedure not only identifies inaccurate computation and
inconsistencies in model implementations but also provides graphical summaries
that can indicate the nature of the problems that arise. We argue that SBC is a
critical part of a robust Bayesian workflow, as well as being a useful tool for
those developing computational algorithms and statistical software.
| stat |
Exact high-dimensional asymptotics for Support Vector Machine | The Support Vector Machine (SVM) is one of the most widely used
classification methods. In this paper, we consider the soft-margin SVM used on
data points with independent features, where the sample size $n$ and the
feature dimension $p$ grows to $\infty$ in a fixed ratio $p/n\rightarrow
\delta$. We propose a set of equations that exactly characterizes the
asymptotic behavior of support vector machine. In particular, we give exact
formulas for (1) the variability of the optimal coefficients, (2) the
proportion of data points lying on the margin boundary (i.e. number of support
vectors), (3) the final objective function value, and (4) the expected
misclassification error on new data points, which in particular implies the
exact formula for the optimal tuning parameter given a data generating
mechanism. We first establish these formulas in the case where the label
$y\in\{+1,-1\}$ is independent of the feature $x$. Then the results are
generalized to the case where the label $y\in\{+1,-1\}$ is allowed to have a
general dependence on the feature $x$ through a linear combination $a_0^Tx$.
These formulas for the non-smooth hinge loss are analogous to the recent
results in \citep{sur2018modern} for smooth logistic loss. Our approach is
based on heuristic leave-one-out calculations.
| stat |
Selection Induced Contrast Estimate (SICE) Effect: An Attempt to
Quantify the Impact of Some Patient Selection Criteria in Randomized Clinical
Trials | Defining the Inclusion/Exclusion (I/E) criteria of a trial is one of the most
important steps during a trial design. Increasingly complex I/E criteria
potentially create information imbalance and transparency issues between the
people who design and run the trials and those who consume the information
produced by the trials. In order to better understand and quantify the impact
of a category of I/E criteria on observed treatment effects, a concept, named
the Selection Induced Contrast Estimate (SICE) effect, is introduced and
formulated in this paper. The SICE effect can exist in controlled clinical
trials when treatment affects the correlation between a marker used for
selection and the response of interest. This effect is demonstrated with both
simulations and real clinical trial data. Although the statistical elements
behind the SICE effect have been well studied, explicitly formulating and
studying this effect can benefit several areas, including better transparency
in I/E criteria, meta-analysis of multiple clinical trials, treatment effect
interpretation in real-world medical practice, etc.
| stat |
Exponential inequalities for nonstationary Markov Chains | Exponential inequalities are main tools in machine learning theory. To prove
exponential inequalities for non i.i.d random variables allows to extend many
learning techniques to these variables. Indeed, much work has been done both on
inequalities and learning theory for time series, in the past 15 years.
However, for the non independent case, almost all the results concern
stationary time series. This excludes many important applications: for example
any series with a periodic behavior is non-stationary. In this paper, we extend
the basic tools of Dedecker and Fan (2015) to nonstationary Markov chains. As
an application, we provide a Bernstein-type inequality, and we deduce risk
bounds for the prediction of periodic autoregressive processes with an unknown
period.
| stat |
FACT: Fast closed testing for exchangeable local tests | Multiple hypothesis testing problems arise naturally in science. In this
paper, we introduce the new Fast Closed Testing (FACT) method for multiple
testing, controlling the family-wise error rate. This error rate is state of
the art in many important application areas, and is preferred to false
discovery rate control for many reasons, including that it leads to stronger
reproducibility. The closure principle rejects an individual hypothesis if all
global nulls of subsets containing it are rejected using some test statistics.
It takes exponential time in the worst case. When the tests are symmetric and
monotone, our method is an exact algorithm for computing the closure, quadratic
in the number of tests, and linear in the number of discoveries. Our framework
generalizes most examples of closed testing such as Holm's and the Bonferroni
method. As a special case of our method, we propose the Simes-higher criticism
fusion test, which is powerful for detecting both a few strong signals, and
also many moderate signals.
| stat |
Asymptotically exact data augmentation: models, properties and
algorithms | Data augmentation, by the introduction of auxiliary variables, has become an
ubiquitous technique to improve convergence properties, simplify the
implementation or reduce the computational time of inference methods such as
Markov chain Monte Carlo ones. Nonetheless, introducing appropriate auxiliary
variables while preserving the initial target probability distribution and
offering a computationally efficient inference cannot be conducted in a
systematic way. To deal with such issues, this paper studies a unified
framework, coined asymptotically exact data augmentation (AXDA), which
encompasses both well-established and more recent approximate augmented models.
In a broader perspective, this paper shows that AXDA models can benefit from
interesting statistical properties and yield efficient inference algorithms. In
non-asymptotic settings, the quality of the proposed approximation is assessed
with several theoretical results. The latter are illustrated on standard
statistical problems. Supplementary materials including computer code for this
paper are available online.
| stat |
Modeling and Probababilistic Forecasting of Natural Gas Prices | In this paper, we examine the problem of modeling and forecasting European
Day-Ahead and Month-Ahead natural gas prices. For this, we propose two distinct
probabilistic models that can be utilized in risk- and portfolio management. We
use daily pricing data ranging from 2011 to 2020. Extensive descriptive data
analysis shows that both time series feature heavy tails, conditional
heteroscedasticity, and show asymmetric behavior in their differences. We
propose state-space time series models under skewed, heavy-tailed distribution
to capture all stylized facts in the data. They include the impact of
autocorrelation, seasonality, risk premia, temperature, storage levels, the
price of European Emission Allowances, and related fuel prices of oil, coal,
and electricity. We provide a rigorous model diagnostic and interpret all model
components in detail. Additionally, we conduct a probabilistic forecasting
study with significance test and compare the predictive performance against
literature benchmarks. The proposed Day-Ahead (Month-Ahead) model leads to a
$13\%$ ($9$\%) reduction in out of sample CRPS compared to the best performing
benchmark model, mainly due to adequate modeling of the volatility and heavy
tails.
| stat |
Bivariate Analysis of Birth Weight and Gestational Age Depending on
Environmental Exposures: Bayesian Distributional Regression with Copulas | In this article, we analyze perinatal data with birth weight (BW) as
primarily interesting response variable. Gestational age (GA) is usually an
important covariate and included in polynomial form. However, in opposition to
this univariate regression, bivariate modeling of BW and GA is recommended to
distinguish effects on each, on both, and between them. Rather than a
parametric bivariate distribution, we apply conditional copula regression,
where marginal distributions of BW and GA (not necessarily of the same form)
can be estimated independently, and where the dependence structure is modeled
conditional on the covariates separately from these marginals. In the resulting
distributional regression models, all parameters of the two marginals and the
copula parameter are observation-specific. Besides biometric and obstetric
information, data on drinking water contamination and maternal smoking are
included as environmental covariates. While the Gaussian distribution is
suitable for BW, the skewed GA data are better modeled by the three-parametric
Dagum distribution. The Clayton copula performs better than the Gumbel and the
symmetric Gaussian copula, indicating lower tail dependence (stronger
dependence when both variables are low), although this non-linear dependence
between BW and GA is surprisingly weak and only influenced by Cesarean section.
A non-linear trend of BW on GA is detected by a classical univariate model that
is polynomial with respect to the effect of GA. Linear effects on BW mean are
similar in both models, while our distributional copula regression also reveals
covariates' effects on all other parameters.
| stat |
Bayesian Computation in Dynamic Latent Factor Models | Bayesian computation for filtering and forecasting analysis is developed for
a broad class of dynamic models. The ability to scale-up such analyses in
non-Gaussian, nonlinear multivariate time series models is advanced through the
introduction of a novel copula construction in sequential filtering of coupled
sets of dynamic generalized linear models. The new copula approach is
integrated into recently introduced multiscale models in which univariate time
series are coupled via nonlinear forms involving dynamic latent factors
representing cross-series relationships. The resulting methodology offers
dramatic speed-up in online Bayesian computations for sequential filtering and
forecasting in this broad, flexible class of multivariate models. Two examples
in nonlinear models for very heterogeneous time series of non-negative counts
demonstrate massive computational efficiencies relative to existing
simulation-based methods, while defining similar filtering and forecasting
outcomes.
| stat |
Quantitative stability of optimal transport maps and linearization of
the 2-Wasserstein space | This work studies an explicit embedding of the set of probability measures
into a Hilbert space, defined using optimal transport maps from a reference
probability density. This embedding linearizes to some extent the 2-Wasserstein
space, and enables the direct use of generic supervised and unsupervised
learning algorithms on measure data. Our main result is that the embedding is
(bi-)H\"older continuous, when the reference density is uniform over a convex
set, and can be equivalently phrased as a dimension-independent
H\"older-stability results for optimal transport maps.
| stat |
Regional economic convergence and spatial quantile regression | The presence of \b{eta}-convergence in European regions is an important issue
to be analyzed. In this paper, we adopt a quantile regression approach in
analyzing economic convergence. While previous work has performed quantile
regression at the national level, we focus on 187 European NUTS2 regions for
the period 1981-2009 and use spatial quantile regression to account for spatial
dependence.
| stat |
A database of travel-related behaviors and attitudes before, during, and
after COVID-19 in the United States | The COVID-19 pandemic has impacted billions of people around the world. To
capture some of these impacts in the United States, we are conducting a
nationwide longitudinal survey collecting information about travel-related
behaviors and attitudes before, during, and after the COVID-19 pandemic. The
survey questions cover a wide range of topics including commuting, daily
travel, air travel, working from home, online learning, shopping, and risk
perception, along with attitudinal, socioeconomic, and demographic information.
Version 1.0 of the survey contains 8,723 responses that are publicly available.
The survey is deployed over multiple waves to the same respondents to monitor
how behaviors and attitudes evolve over time. This article details the
methodology adopted for the collection, cleaning, and processing of the data.
In addition, the data are weighted to be representative of national and
regional demographics. This survey dataset can aid researchers, policymakers,
businesses, and government agencies in understanding both the extent of
behavioral shifts and the likelihood that these changes will persist after
COVID-19.
| stat |
Parametric and non-parametric estimation of extreme earthquake event:
the joint tail inference for mainshocks and aftershocks | In an earthquake event, the combination of a strong mainshock and damaging
aftershocks is often the cause of severe structural damages and/or high death
tolls. The objective of this paper is to provide estimation for the probability
of such extreme events where the mainshock and the largest aftershocks exceed
certain thresholds. Two approaches are illustrated and compared -- a parametric
approach based on previously observed stochastic laws in earthquake data, and a
non-parametric approach based on bivariate extreme value theory. We analyze the
earthquake data from the North Anatolian Fault Zone (NAFZ) in Turkey during
1965-2018 and show that the two approaches provide unifying results.
| stat |
Substitutes for the non-existent square lattice designs for 36 varieties | Square lattice designs are often used in trials of new varieties of various
agricultural crops. However, there are no square lattice designs for 36
varieties in blocks of size six for four or more replicates. Here we use three
different approaches to construct designs for up to eight replicates. All the
designs perform well in terms of giving a low average variance of variety
contrasts.
Supplementary materials are available online.
| stat |
Online Regularization for High-Dimensional Dynamic Pricing Algorithms | We propose a novel \textit{online regularization} scheme for
revenue-maximization in high-dimensional dynamic pricing algorithms. The online
regularization scheme equips the proposed optimistic online regularized maximum
likelihood pricing (\texttt{OORMLP}) algorithm with three major advantages:
encode market noise knowledge into pricing process optimism; empower online
statistical learning with always-validity over all decision points; envelop
prediction error process with time-uniform non-asymptotic oracle inequalities.
This type of non-asymptotic inference results allows us to design safer and
more robust dynamic pricing algorithms in practice. In theory, the proposed
\texttt{OORMLP} algorithm exploits the sparsity structure of high-dimensional
models and obtains a logarithmic regret in a decision horizon. These
theoretical advances are made possible by proposing an optimistic online LASSO
procedure that resolves dynamic pricing problems at the \textit{process} level,
based on a novel use of non-asymptotic martingale concentration. In
experiments, we evaluate \texttt{OORMLP} in different synthetic pricing problem
settings and observe that \texttt{OORMLP} performs better than \texttt{RMLP}
proposed in \cite{javanmard2019dynamic}.
| stat |
Comparison of Bayesian Nonparametric Density Estimation Methods | In this paper, we propose a nonparametric Bayesian approach for Lindsey and
penalized Gaussian mixtures methods. We compare these methods with the
Dirichlet process mixture model. Our approach is a Bayesian nonparametric
method not based solely on a parametric family of probability distributions.
Thus, the fitted models are more robust to model misspecification. Also, with
the Bayesian approach, we have the entire posterior distribution of our
parameter of interest; it can be summarized through credible intervals, mean,
median, standard deviation, quantiles, etc. The Lindsey, penalized Gaussian
mixtures, and Dirichlet process mixture methods are reviewed. The estimations
are performed via Markov chain Monte Carlo (MCMC) methods. The penalized
Gaussian mixtures method is implemented via Hamiltonian Monte Carlo (HMC). We
show that under certain regularity conditions, and as n increases, the
posterior distribution of the weights converges to a Normal distribution.
Simulation results and data analysis are reported.
| stat |
Covariance-Controlled Adaptive Langevin Thermostat for Large-Scale
Bayesian Sampling | Monte Carlo sampling for Bayesian posterior inference is a common approach
used in machine learning. The Markov Chain Monte Carlo procedures that are used
are often discrete-time analogues of associated stochastic differential
equations (SDEs). These SDEs are guaranteed to leave invariant the required
posterior distribution. An area of current research addresses the computational
benefits of stochastic gradient methods in this setting. Existing techniques
rely on estimating the variance or covariance of the subsampling error, and
typically assume constant variance. In this article, we propose a
covariance-controlled adaptive Langevin thermostat that can effectively
dissipate parameter-dependent noise while maintaining a desired target
distribution. The proposed method achieves a substantial speedup over popular
alternative schemes for large-scale machine learning applications.
| stat |
Geometric mean extension for data sets with zeros | There are numerous examples in different research fields where the use of the
geometric mean is more appropriate than the arithmetic mean. However, the
geometric mean has a serious limitation in comparison with the arithmetic mean.
Means are used to summarize the information in a large set of values in a
single number; yet, the geometric mean of a data set with at least one zero is
always zero. As a result, the geometric mean does not capture any information
about the non-zero values. The purpose of this short contribution is to review
solutions proposed in the literature that enable the computation of the
geometric mean of data sets containing zeros and to show that they do not
fulfil the `recovery' or `monotonicity' conditions that we define. The standard
geometric mean should be recovered from the modified geometric mean if the data
set does not contain any zeros (recovery condition). Also, if the values of an
ordered data set are greater one by one than the values of another data set
then the modified geometric mean of the first data set must be greater than the
modified geometric mean of the second data set (monotonicity condition). We
then formulate a modified version of the geometric mean that can handle zeros
while satisfying both desired conditions.
| stat |
Sparse Gaussian Process Variational Autoencoders | Large, multi-dimensional spatio-temporal datasets are omnipresent in modern
science and engineering. An effective framework for handling such data are
Gaussian process deep generative models (GP-DGMs), which employ GP priors over
the latent variables of DGMs. Existing approaches for performing inference in
GP-DGMs do not support sparse GP approximations based on inducing points, which
are essential for the computational efficiency of GPs, nor do they handle
missing data -- a natural occurrence in many spatio-temporal datasets -- in a
principled manner. We address these shortcomings with the development of the
sparse Gaussian process variational autoencoder (SGP-VAE), characterised by the
use of partial inference networks for parameterising sparse GP approximations.
Leveraging the benefits of amortised variational inference, the SGP-VAE enables
inference in multi-output sparse GPs on previously unobserved data with no
additional training. The SGP-VAE is evaluated in a variety of experiments where
it outperforms alternative approaches including multi-output GPs and structured
VAEs.
| stat |
Fairness Assessment for Artificial Intelligence in Financial Industry | Artificial Intelligence (AI) is an important driving force for the
development and transformation of the financial industry. However, with the
fast-evolving AI technology and application, unintentional bias, insufficient
model validation, immature contingency plan and other underestimated threats
may expose the company to operational and reputational risks. In this paper, we
focus on fairness evaluation, one of the key components of AI Governance,
through a quantitative lens. Statistical methods are reviewed for imbalanced
data treatment and bias mitigation. These methods and fairness evaluation
metrics are then applied to a credit card default payment example.
| stat |
Behavior Associations in Lone-Actor Terrorists | Terrorist attacks carried out by individuals or single cells have
significantly accelerated over the last 20 years. This type of terrorism,
defined as lone-actor (LA) terrorism, stands as one of the greatest security
threats of our time. Research on LA behavior and characteristics has emerged
and accelerated over the last decade. While these studies have produced
valuable information on demographics, behavior, classifications, and warning
signs, the relationship among these characters are yet to be addressed.
Moreover, the means of radicalization and attacking have changed over decades.
This study first identifies 25 binary behavioral characteristics of LAs and
analyzes 192 LAs recorded on three different databases. Next, the
classification is carried out according to first ideology, then to incident
scene behavior via a virtual attacker-defender game, and, finally, according to
the clusters obtained from the data. In addition, within each class,
statistically significant associations and temporal relations are extracted
using the A-priori algorithm. These associations would be instrumental in
identifying the attacker type and intervene at the right time. The results
indicate that while pre-9/11 LAs were mostly radicalized by the people in their
environment, post-9/11 LAs are more diverse. Furthermore, the association
chains for different LA types present unique characteristic pathways to
violence and after-attack behavior.
| stat |
Comprehensive Privacy Analysis of Deep Learning: Passive and Active
White-box Inference Attacks against Centralized and Federated Learning | Deep neural networks are susceptible to various inference attacks as they
remember information about their training data. We design white-box inference
attacks to perform a comprehensive privacy analysis of deep learning models. We
measure the privacy leakage through parameters of fully trained models as well
as the parameter updates of models during training. We design inference
algorithms for both centralized and federated learning, with respect to passive
and active inference attackers, and assuming different adversary prior
knowledge.
We evaluate our novel white-box membership inference attacks against deep
learning algorithms to trace their training data records. We show that a
straightforward extension of the known black-box attacks to the white-box
setting (through analyzing the outputs of activation functions) is ineffective.
We therefore design new algorithms tailored to the white-box setting by
exploiting the privacy vulnerabilities of the stochastic gradient descent
algorithm, which is the algorithm used to train deep neural networks. We
investigate the reasons why deep learning models may leak information about
their training data. We then show that even well-generalized models are
significantly susceptible to white-box membership inference attacks, by
analyzing state-of-the-art pre-trained and publicly available models for the
CIFAR dataset. We also show how adversarial participants, in the federated
learning setting, can successfully run active membership inference attacks
against other participants, even when the global model achieves high prediction
accuracies.
| stat |
Probabilistic Detection and Estimation of Conic Sections from Noisy Data | Inferring unknown conic sections on the basis of noisy data is a challenging
problem with applications in computer vision. A major limitation of the
currently available methods for conic sections is that estimation methods rely
on the underlying shape of the conics (being known to be ellipse, parabola or
hyperbola). A general purpose Bayesian hierarchical model is proposed for conic
sections and corresponding estimation method based on noisy data is shown to
work even when the specific nature of the conic section is unknown. The model,
thus, provides probabilistic detection of the underlying conic section and
inference about the associated parameters of the conic section. Through
extensive simulation studies where the true conics may not be known, the
methodology is demonstrated to have practical and methodological advantages
relative to many existing techniques. In addition, the proposed method provides
probabilistic measures of uncertainty of the estimated parameters. Furthermore,
we observe high fidelity to the true conics even in challenging situations,
such as data arising from partial conics in arbitrarily rotated and
non-standard form, and where a visual inspection is unable to correctly
identify the type of conic section underlying the data.
| stat |
The SIR-P Model: An Illustration of the Screening Paradox | In previous work by this author, the screening paradox - the loss of
predictive power of screening tests over time $t$ - was mathematically
formalized using Bayesian theory. Where $J$ is Youden's statistic, $b$ is the
specificity of the screening test and $\phi$ is the prevalence of disease, the
ratio of positive predictive values at subsequent time $k$, $\rho(\phi_{k})$,
over the original $\rho(\phi_{0})$ at $t_0$ is given by:
$\zeta(\phi_{0},k) = \frac{\rho(\phi_{k})}{\rho(\phi_{0})}
=\frac{\phi_k(1-b)+J\phi_0\phi_k}{\phi_0(1-b)+J\phi_0\phi_k}$
Herein, we modify the traditional Kermack-McKendrick SIR Model to include the
fluctuation of the positive predictive value $\rho(\phi)$ (PPV) of a screening
test over time as a function of the prevalence threshold $\phi_e$. We term this
modified model the SIR-P model. Where a = sensitivity, b = specificity, $S$ =
number susceptible, $I$ = number infected, $R$ = number recovered/dead, $\beta$
= infectious rate, $\gamma$ = recovery rate, and $N$ is the total number in the
population, the predictive value $\rho(\phi,t)$ over time $t$ is given by:
$\rho(\phi,t) = \frac{a[\frac{\beta IS}{N}-\gamma I]}{ a[\frac{\beta
IS}{N}-\gamma I]+(1-b)(1-[\frac{\beta IS}{N}-\gamma I])}$
Otherwise stated:
$\rho(\phi,t) = \frac{a\frac{dI}{dt}}{
a\frac{dI}{dt}+(1-b)(1-\frac{dI}{dt})}$ where $\frac{dI}{dt}$ is the
fluctuation of infected individuals over time $t$.
| stat |
Protocol for an Observational Study of the Association of High School
Football Participation on Health in Late Adulthood | American football is the most popular high school sport and is among the
leading cause of injury among adolescents. While there has been considerable
recent attention on the link between football and cognitive decline, there is
also evidence of higher than expected rates of pain, obesity, and lower quality
of life among former professional players, either as a result of repetitive
head injury or through different mechanisms. Previously hidden downstream
effects of playing football may have far-reaching public health implications
for participants in youth and high school football programs.
Our proposed study is a retrospective observational study that compares 1,153
high school males who played varsity football with 2,751 male students who did
not. 1,951 of the control subjects did not play any sport and the remaining 800
controls played a non-contact sport. Our primary outcome is self-rated health
measured at age 65. To control for potential confounders, we adjust for
pre-exposure covariates with matching and model-based covariance adjustment. We
will conduct an ordered testing procedure designed to use the full pool of
2,751 controls while also controlling for possible unmeasured differences
between students who played sports and those who did not. We will
quantitatively assess the sensitivity of the results to potential unmeasured
confounding. The study will also assess secondary outcomes of pain, difficulty
with activities of daily living, and obesity, as these are both important to
individual well-being and have public health relevance.
| stat |
Efficient and Robust Propensity-Score-Based Methods for Population
Inference using Epidemiologic Cohorts | Most epidemiologic cohorts are composed of volunteers who do not represent
the general population. To enable population inference from cohorts, we and
others have proposed utilizing probability survey samples as external
references to develop a propensity score (PS) for membership in the cohort
versus survey. Herein we develop a unified framework for PS-based weighting
(such as inverse PS weighting (IPSW)) and matching methods (such as
kernel-weighting (KW) method). We identify a fundamental Strong Exchangeability
Assumption (SEA) underlying existing PS-based matching methods whose failure
invalidates inference even if the PS-model is correctly specified. We relax the
SEA to a Weak Exchangeability Assumption (WEA) for the matching method. Also,
we propose IPSW.S and KW.S methods that reduce the variance of PS-based
estimators by scaling the survey weights used in the PS estimation. We prove
consistency of the IPSW.S and KW.S estimators of population means and
prevalences under WEA, and provide asymptotic variances and consistent variance
estimators. In simulations, the KW.S and IPSW.S estimators had smallest MSE. In
our data example, the original KW estimates had large bias, whereas the KW.S
estimates had the smallest MSE.
| stat |
Learning Latent Dynamics for Partially-Observed Chaotic Systems | This paper addresses the data-driven identification of latent dynamical
representations of partially-observed systems, i.e., dynamical systems for
which some components are never observed, with an emphasis on forecasting
applications, including long-term asymptotic patterns. Whereas state-of-the-art
data-driven approaches rely on delay embeddings and linear decompositions of
the underlying operators, we introduce a framework based on the data-driven
identification of an augmented state-space model using a neural-network-based
representation. For a given training dataset, it amounts to jointly learn an
ODE (Ordinary Differential Equation) representation in the latent space and
reconstructing latent states. Through numerical experiments, we demonstrate the
relevance of the proposed framework w.r.t. state-of-the-art approaches in terms
of short-term forecasting performance and long-term behaviour. We further
discuss how the proposed framework relates to Koopman operator theory and
Takens' embedding theorem.
| stat |
Kernel Ridge Regression with kernel Laplacian Regularization | Laplacian regularization is a popular smoothing technique in machine
learning. It is particularly useful in situations where ambiguity of the data
imposes the use of a criterion to disambiguate between potential functions
explaining data, could it be spectral clustering or semi-supervised learning.
While Laplacian regularization is usually approached through neighborhood graph
diffusion, we present an approach through kernel methods, based on derivative
evaluation maps. We derive an analytical solution of the empirical risk
minimization with kernel Laplacian regularization. We prove strong consistency
of our estimate when the number of data goes towards infinity. Moreover, we
show that, under regularity assumptions, our kernel method bypasses the curse
of dimensionality, hence providing a strong alternative to neighborhood graph
methods that do not avoid it.
| stat |
Conditionally Gaussian Random Sequences for an Integrated Variance
Estimator with Correlation between Noise and Returns | Correlation between microstructure noise and latent financial logarithmic
returns is an empirically relevant phenomenon with sound theoretical
justification. With few notable exceptions, all integrated variance estimators
proposed in the financial literature are not designed to explicitly handle such
a dependence, or handle it only in special settings. We provide an integrated
variance estimator that is robust to correlated noise and returns. For this
purpose, a generalization of the Forward Filtering Backward Sampling algorithm
is proposed, to provide a sampling technique for a latent conditionally
Gaussian random sequence. We apply our methodology to intra-day Microsoft
prices, and compare it in a simulation study with established alternatives,
showing an advantage in terms of root mean square error and dispersion.
| stat |
Building a COVID-19 Vulnerability Index | COVID-19 is an acute respiratory disease that has been classified as a
pandemic by the World Health Organization. Characterization of this disease is
still in its early stages. However, it is known to have high mortality rates,
particularly among individuals with preexisting medical conditions. Creating
models to identify individuals who are at the greatest risk for severe
complications due to COVID-19 will be useful for outreach campaigns to help
mitigate the disease's worst effects. While information specific to COVID-19 is
limited, a model using complications due to other upper respiratory infections
can be used as a proxy to help identify those individuals who are at the
greatest risk. We present the results for three models predicting such
complications, with each model increasing predictive effectiveness at the
expense of ease of implementation.
| stat |
Model-Independent Detection of New Physics Signals Using Interpretable
Semi-Supervised Classifier Tests | A central goal in experimental high energy physics is to detect new physics
signals that are not explained by known physics. In this paper, we aim to
search for new signals that appear as deviations from known Standard Model
physics in high-dimensional particle physics data. To do this, we determine
whether there is any statistically significant difference between the
distribution of Standard Model background samples and the distribution of the
experimental observations, which are a mixture of the background and a
potential new signal. Traditionally, one also assumes access to a sample from a
model for the hypothesized signal distribution. Here we instead investigate a
model-independent method that does not make any assumptions about the signal
and uses a semi-supervised classifier to detect the presence of the signal in
the experimental data. We construct three test statistics using the classifier:
an estimated likelihood ratio test (LRT) statistic, a test based on the area
under the ROC curve (AUC), and a test based on the misclassification error
(MCE). Additionally, we propose a method for estimating the signal strength
parameter and explore active subspace methods to interpret the proposed
semi-supervised classifier in order to understand the properties of the
detected signal. We investigate the performance of the methods on a data set
related to the search for the Higgs boson at the Large Hadron Collider at CERN.
We demonstrate that the semi-supervised tests have power competitive with the
classical supervised methods for a well-specified signal, but much higher power
for an unexpected signal which might be entirely missed by the supervised
tests.
| stat |
B-Value and Empirical Equivalence Bound: A New Procedure of Hypothesis
Testing | In this study, we propose a two-stage procedure for hypothesis testing, where
the first stage is conventional hypothesis testing and the second is an
equivalence testing procedure using an introduced Empirical Equivalence Bound.
In 2016, the American Statistical Association released a policy statement on
P-values to clarify the proper use and interpretation in response to the
criticism of reproducibility and replicability in scientific findings. A recent
solution to improve reproducibility and transparency in statistical hypothesis
testing is to integrate P-values (or confidence intervals) with practical or
scientific significance. Similar ideas have been proposed via the equivalence
test, where the goal is to infer equality under a presumption (null) of
inequality of parameters. However, in these testing procedures, the definition
of scientific significance/equivalence can be subjective. To circumvent this
drawback, we introduce a B-value and the Empirical Equivalence Bound, which are
both estimated from the data. Performing a second-stage equivalence test, our
procedure offers an opportunity to correct for false positive discoveries and
improve the reproducibility in findings across studies.
| stat |
Does hospital cooperation increase the quality of healthcare? | Motivated by reasons such as altruism, managers from different hospitals may
engage in cooperative behaviours, which shape the networked healthcare economy.
In this paper we study the determinants of hospital cooperation and its
association with the quality delivered by hospitals, using Italian
administrative data. We explore the impact on patient transfers between
hospitals (cooperation/network) of a set of demand-supply factors, as well as
distance-based centrality measures. We then use this framework to assess how
such cooperation is related to the overall quality for the hospital of origin
and of destination of the patient transfer. The over-dispersed Poisson mixed
model that we propose, inspired by the literature on social relations models,
is suitably defined to handle network data, which are rarely used in health
economics. The results show that distance plays an important role in hospital
cooperation, though there are other factors that matter such as geographical
centrality. Another empirical finding is the existence of a positive
relationship between hospital cooperation and the overall quality of the
connected hospitals. The absence of a source of information on the quality of
hospitals accessible to all providers, such as in the form of star ratings, may
prevent some hospitals to engage and cooperate with other hospitals of
potentially higher quality. This may result in a lower degree of cooperation
among hospitals and a reduction in quality overall.
| stat |
Scalable Online Survey Framework: from Sampling to Analysis | With the advancement in technology, raw event data generated by the digital
world have grown tremendously. However, such data tend to be insufficient and
noisy when it comes to measuring user intention or satisfaction. One effective
way to measure user experience directly is through surveys. In particular, with
the popularity of online surveys, extensive work has been put in to study this
field. Surveys at LinkedIn play a major role in influencing product and
marketing decisions and supporting our sales efforts. We run an increasing
number of surveys that help us understand shifts in awareness and perceptions
with regards to our own products and also to peer companies. As the need to
survey grows, both sampling and analysis of surveys have become more
challenging. Instead of simply multiplying the number of surveys each user
takes, we need a scalable approach to collect enough and representative samples
for each survey analysis while maintaining good user experience. In this paper,
we start with discussions on how we handle multiple email surveys under such
constraints. We then shift our discussions to challenges of in-product surveys
and how we address them at LinkedIn through a survey study conducted across two
mobile apps. Finally, we share how in-product surveys can be utilized as
monitoring tools and connect surveys with A/B testing.
| stat |
On Target Shift in Adversarial Domain Adaptation | Discrepancy between training and testing domains is a fundamental problem in
the generalization of machine learning techniques. Recently, several approaches
have been proposed to learn domain invariant feature representations through
adversarial deep learning. However, label shift, where the percentage of data
in each class is different between domains, has received less attention. Label
shift naturally arises in many contexts, especially in behavioral studies where
the behaviors are freely chosen. In this work, we propose a method called
Domain Adversarial nets for Target Shift (DATS) to address label shift while
learning a domain invariant representation. This is accomplished by using
distribution matching to estimate label proportions in a blind test set. We
extend this framework to handle multiple domains by developing a scheme to
upweight source domains most similar to the target domain. Empirical results
show that this framework performs well under large label shift in synthetic and
real experiments, demonstrating the practical importance.
| stat |
Diagnosing model misspecification and performing generalized Bayes'
updates via probabilistic classifiers | Model misspecification is a long-standing enigma of the Bayesian inference
framework as posteriors tend to get overly concentrated on ill-informed
parameter values towards the large sample limit. Tempering of the likelihood
has been established as a safer way to do updates from prior to posterior in
the presence of model misspecification. At one extreme tempering can ignore the
data altogether and at the other extreme it provides the standard Bayes' update
when no misspecification is assumed to be present. However, it is an open issue
how to best recognize misspecification and choose a suitable level of tempering
without access to the true generating model. Here we show how probabilistic
classifiers can be employed to resolve this issue. By training a probabilistic
classifier to discriminate between simulated and observed data provides an
estimate of the ratio between the model likelihood and the likelihood of the
data under the unobserved true generative process, within the discriminatory
abilities of the classifier. The expectation of the logarithm of a ratio with
respect to the data generating process gives an estimation of the negative
Kullback-Leibler divergence between the statistical generative model and the
true generative distribution. Using a set of canonical examples we show that
this divergence provides a useful misspecification diagnostic, a model
comparison tool, and a method to inform a generalised Bayesian update in the
presence of misspecification for likelihood-based models.
| stat |
Efficient Path Algorithms for Clustered Lasso and OSCAR | In high dimensional regression, feature clustering by their effects on
outcomes is often as important as feature selection. For that purpose,
clustered Lasso and octagonal shrinkage and clustering algorithm for regression
(OSCAR) are used to make feature groups automatically by pairwise $L_1$ norm
and pairwise $L_\infty$ norm, respectively. This paper proposes efficient path
algorithms for clustered Lasso and OSCAR to construct solution paths with
respect to their regularization parameters. Despite too many terms in
exhaustive pairwise regularization, their computational costs are reduced by
using symmetry of those terms. Simple equivalent conditions to check
subgradient equations in each feature group are derived by some graph theories.
The proposed algorithms are shown to be more efficient than existing algorithms
in numerical experiments.
| stat |
Designing a statistical procedure for monitoring global carbon dioxide
emissions | Following the Paris Agreement of $2015$, most countries have agreed to reduce
their carbon dioxide (CO$_2$) emissions according to individually set
Nationally Determined Contributions. However, national CO$_2$ emissions are
reported by individual countries and cannot be directly measured or verified by
third parties. Inherent weaknesses in the reporting methodology may
misrepresent, typically an under-reporting of, the total national emissions.
This paper applies the theory of sequential testing to design a statistical
monitoring procedure that can be used to detect systematic under-reportings of
CO$_2$ emissions. Using simulations, we investigate how the proposed sequential
testing procedure can be expected to work in practice. We find that, if
emissions are reported faithfully, the test is correctly sized, while, if
emissions are under-reported, detection time can be sufficiently fast to help
inform the $5$ yearly global "stocktake" of the Paris Agreement. We recommend
the monitoring procedure be applied going forward as part of a larger portfolio
of methods designed to verify future global CO$_2$ emissions.
| stat |
Embedding and learning with signatures | Sequential and temporal data arise in many fields of research, such as
quantitative finance, medicine, or computer vision. A novel approach for
sequential learning, called the signature method and rooted in rough path
theory, is considered. Its basic principle is to represent multidimensional
paths by a graded feature set of their iterated integrals, called the
signature. This approach relies critically on an embedding principle, which
consists in representing discretely sampled data as paths, i.e., functions from
$[0,1]$ to $\mathbb{R}^d$. After a survey of machine learning methodologies for
signatures, the influence of embeddings on prediction accuracy is investigated
with an in-depth study of three recent and challenging datasets. It is shown
that a specific embedding, called lead-lag, is systematically the strongest
performer across all datasets and algorithms considered. Moreover, an empirical
study reveals that computing signatures over the whole path domain does not
lead to a loss of local information. It is concluded that, with a good
embedding, combining signatures with other simple algorithms achieves results
competitive with state-of-the-art, domain-specific approaches.
| stat |
Stochastic Gradient MCMC for State Space Models | State space models (SSMs) are a flexible approach to modeling complex time
series. However, inference in SSMs is often computationally prohibitive for
long time series. Stochastic gradient MCMC (SGMCMC) is a popular method for
scalable Bayesian inference for large independent data. Unfortunately when
applied to dependent data, such as in SSMs, SGMCMC's stochastic gradient
estimates are biased as they break crucial temporal dependencies. To alleviate
this, we propose stochastic gradient estimators that control this bias by
performing additional computation in a `buffer' to reduce breaking
dependencies. Furthermore, we derive error bounds for this bias and show a
geometric decay under mild conditions. Using these estimators, we develop novel
SGMCMC samplers for discrete, continuous and mixed-type SSMs with analytic
message passing. Our experiments on real and synthetic data demonstrate the
effectiveness of our SGMCMC algorithms compared to batch MCMC, allowing us to
scale inference to long time series with millions of time points.
| stat |
Identification of high-energy astrophysical point sources via
hierarchical Bayesian nonparametric clustering | The light we receive from distant astrophysical objects carries information
about their origins and the physical mechanisms that power them. The study of
these signals, however, is complicated by the fact that observations are often
a mixture of the light emitted by multiple localized sources situated in a
spatially-varying background. A general algorithm to achieve robust and
accurate source identification in this case remains an open question in
astrophysics.
This paper focuses on high-energy light (such as X-rays and gamma-rays), for
which observatories can detect individual photons (quanta of light), measuring
their incoming direction, arrival time, and energy. Our proposed Bayesian
methodology uses both the spatial and energy information to identify point
sources, that is, separate them from the spatially-varying background, to
estimate their number, and to compute the posterior probabilities that each
photon originated from each identified source. This is accomplished via a
Dirichlet process mixture while the background is simultaneously reconstructed
via a flexible Bayesian nonparametric model based on B-splines. Our proposed
method is validated with a suite of simulation studies and illustrated with an
application to a complex region of the sky observed by the \emph{Fermi}
Gamma-ray Space Telescope.
| stat |
A Tool for Custom Construction of QMC and RQMC Point Sets | We present LatNet Builder, a software tool to find good parameters for
lattice rules, polynomial lattice rules, and digital nets in base 2, for
quasi-Monte Carlo (QMC) and randomized quasi-Monte Carlo (RQMC) sampling over
the $s$-dimensional unit hypercube. The selection criteria are figures of merit
that give different weights to different subsets of coordinates. They are upper
bounds on the worst-case error (for QMC) or variance (for RQMC) for integrands
rescaled to have a norm of at most one in certain Hilbert spaces of functions.
Various Hilbert spaces, figures of merit, types of constructions, and search
methods are covered by the tool. We provide simple illustrations of what it can
do.
| stat |
Parallel Bayesian Global Optimization of Expensive Functions | We consider parallel global optimization of derivative-free
expensive-to-evaluate functions, and propose an efficient method based on
stochastic approximation for implementing a conceptual Bayesian optimization
algorithm proposed by Ginsbourger et al. (2007). At the heart of this algorithm
is maximizing the information criterion called the "multi-points expected
improvement'', or the q-EI. To accomplish this, we use infinitessimal
perturbation analysis (IPA) to construct a stochastic gradient estimator and
show that this estimator is unbiased. We also show that the stochastic gradient
ascent algorithm using the constructed gradient estimator converges to a
stationary point of the q-EI surface, and therefore, as the number of multiple
starts of the gradient ascent algorithm and the number of steps for each start
grow large, the one-step Bayes optimal set of points is recovered. We show in
numerical experiments that our method for maximizing the q-EI is faster than
methods based on closed-form evaluation using high-dimensional integration,
when considering many parallel function evaluations, and is comparable in speed
when considering few. We also show that the resulting one-step Bayes optimal
algorithm for parallel global optimization finds high-quality solutions with
fewer evaluations than a heuristic based on approximately maximizing the q-EI.
A high-quality open source implementation of this algorithm is available in the
open source Metrics Optimization Engine (MOE).
| stat |
Clarifying species dependence under joint species distribution modeling | Joint species distribution modeling is attracting increasing attention these
days, acknowledging the fact that individual level modeling fails to take into
account expected dependence/interaction between species. These models attempt
to capture species dependence through an associated correlation matrix arising
from a set of latent multivariate normal variables. However, these associations
offer little insight into dependence behavior between species at sites.
We focus on presence/absence data using joint species modeling which
incorporates spatial dependence between sites. For pairs of species, we
emphasize the induced odds ratios (along with the joint probabilities of
occurrence); they provide much clearer understanding of joint presence/absence
behavior. In fact, we propose a spatial odds ratio surface over the region of
interest to capture how dependence varies over the region.
We illustrate with a dataset from the Cape Floristic Region of South Africa
consisting of more than 600 species at more than 600 sites. We present the
spatial distribution of odds ratios for pairs of species that are positively
correlated and pairs that are negatively correlated under the joint species
distribution model.
The multivariate normal covariance matrix associated with a collection of
species is only a device for creating dependence among species but it lacks
interpretation. By considering odds ratios, the quantitative ecologist will be
able to better appreciate the practical dependence between species that is
implicit in these joint species distribution modeling specifications.
| stat |
A kernel test for quasi-independence | We consider settings in which the data of interest correspond to pairs of
ordered times, e.g, the birth times of the first and second child, the times at
which a new user creates an account and makes the first purchase on a website,
and the entry and survival times of patients in a clinical trial. In these
settings, the two times are not independent (the second occurs after the
first), yet it is still of interest to determine whether there exists
significant dependence {\em beyond} their ordering in time. We refer to this
notion as "quasi-(in)dependence". For instance, in a clinical trial, to avoid
biased selection, we might wish to verify that recruitment times are
quasi-independent of survival times, where dependencies might arise due to
seasonal effects. In this paper, we propose a nonparametric statistical test of
quasi-independence. Our test considers a potentially infinite space of
alternatives, making it suitable for complex data where the nature of the
possible quasi-dependence is not known in advance. Standard parametric
approaches are recovered as special cases, such as the classical conditional
Kendall's tau, and log-rank tests. The tests apply in the right-censored
setting: an essential feature in clinical trials, where patients can withdraw
from the study. We provide an asymptotic analysis of our test-statistic, and
demonstrate in experiments that our test obtains better power than existing
approaches, while being more computationally efficient.
| stat |
Rectangular Bounding Process | Stochastic partition models divide a multi-dimensional space into a number of
rectangular regions, such that the data within each region exhibit certain
types of homogeneity. Due to the nature of their partition strategy, existing
partition models may create many unnecessary divisions in sparse regions when
trying to describe data in dense regions. To avoid this problem we introduce a
new parsimonious partition model -- the Rectangular Bounding Process (RBP) --
to efficiently partition multi-dimensional spaces, by employing a bounding
strategy to enclose data points within rectangular bounding boxes. Unlike
existing approaches, the RBP possesses several attractive theoretical
properties that make it a powerful nonparametric partition prior on a
hypercube. In particular, the RBP is self-consistent and as such can be
directly extended from a finite hypercube to infinite (unbounded) space. We
apply the RBP to regression trees and relational models as a flexible partition
prior. The experimental results validate the merit of the RBP {in rich yet
parsimonious expressiveness} compared to the state-of-the-art methods.
| stat |
Spatio-Temporal Reconstructions of Global CO2-Fluxes using Gaussian
Markov Random Fields | Atmospheric inverse modelling is a method for reconstructing historical
fluxes of green-house gas between land and atmosphere, using observed
atmospheric concentrations and an atmospheric tracer transport model. The small
number of observed atmospheric concentrations in relation to the number of
unknown flux components makes the inverse problem ill-conditioned, and
assumptions on the fluxes are needed to constrain the solution. A common
practise is to model the fluxes using latent Gaussian fields with a mean
structure based on estimated fluxes from combinations of process modelling
(natural fluxes) and statistical bookkeeping (anthropogenic emissions). Here,
we reconstruct global \CO flux fields by modelling fluxes using Gaussian Markov
Random Fields (GMRF), resulting in a flexible and computational beneficial
model with a Mat\'ern-like spatial covariance, and a temporal covariance
defined through an auto-regressive model with seasonal dependence.
In contrast to previous inversions, the flux is defined on a spatially
continuous domain, and the traditionally discrete flux representation is
replaced by integrated fluxes at the resolution specified by the transport
model. This formulation removes aggregation errors in the flux covariance, due
to the traditional representation of area integrals by fluxes at discrete
points, and provides a model closer resembling real-life space-time continuous
fluxes.
| stat |
Extreme expectile estimation for heavy-tailed time series | Expectiles are a least squares analogue of quantiles which have lately
received substantial attention in actuarial and financial risk management
contexts. Unlike quantiles, expectiles define coherent risk measures and are
determined by tail expectations rather than tail probabilities; unlike the
Expected Shortfall, they define elicitable risk measures. This has motivated
recent studies of the behaviour and estimation of extreme expectile-based risk
measures. The case of stationary but weakly dependent observations has,
however, been left largely untouched, even though correctly accounting for the
uncertainty present in typical financial applications requires the
consideration of dependent data. We investigate the estimation of, and
construction of accurate confidence intervals for, extreme expectiles and
expectile-based Marginal Expected Shortfall in a general $\beta-$mixing
context, containing the classes of ARMA, ARCH and GARCH models with
heavy-tailed innovations that are of interest in financial applications. The
methods are showcased in a numerical simulation study and on real financial
data.
| stat |
A Quadratic Programming Solution to the FICO Credit Scoring Problem | After decades of experience in developing credit scores, the FICO corporation
has formulated the FICO Credit Scoring Problem as follows: Find the Generalized
Additive Model (GAM), with component step functions, that maximizes divergence
subject to the PILE (Palatability, Interpretability, Legal, Explain-ability)
constraints. The PILE constraints are also called shape constraints, and
satisfying them is called score engineering. Before 2003, FICO used an
algorithm, based on Linear Programing, to approximately solve the FICO Credit
Scoring Problem. In this paper, I develop an exact solution to the FICO Credit
Scoring Problem. Finding the exact solution has eluded FICO for years.
Divergence is a ratio of quadratic functions of the score weights. I show that
the max divergence problem can be transformed into a quadratic program. The
quadratic programming formulation allows one to handle the PILE constraints
very easily. FICO currently uses aspects of this technology to develop the
famous FICO Credit Score.
| stat |
Coloring Panchromatic Nighttime Satellite Images: Comparing the
Performance of Several Machine Learning Methods | Artificial light-at-night (ALAN), emitted from the ground and visible from
space, marks human presence on Earth. Since the launch of the Suomi National
Polar Partnership satellite with the Visible Infrared Imaging Radiometer Suite
Day/Night Band (VIIRS/DNB) onboard, global nighttime images have significantly
improved; however, they remained panchromatic. Although multispectral images
are also available, they are either commercial or free of charge, but sporadic.
In this paper, we use several machine learning techniques, such as linear,
kernel, random forest regressions, and elastic map approach, to transform
panchromatic VIIRS/DBN into Red Green Blue (RGB) images. To validate the
proposed approach, we analyze RGB images for eight urban areas worldwide. We
link RGB values, obtained from ISS photographs, to panchromatic ALAN
intensities, their pixel-wise differences, and several land-use type proxies.
Each dataset is used for model training, while other datasets are used for the
model validation. The analysis shows that model-estimated RGB images
demonstrate a high degree of correspondence with the original RGB images from
the ISS database. Yet, estimates, based on linear, kernel and random forest
regressions, provide better correlations, contrast similarity and lower WMSEs
levels, while RGB images, generated using elastic map approach, provide higher
consistency of predictions.
| stat |
Bayesian Model Calibration for Extrapolative Prediction via Gibbs
Posteriors | The current standard Bayesian approach to model calibration, which assigns a
Gaussian process prior to the discrepancy term, often suffers from issues of
unidentifiability and computational complexity and instability. When the goal
is to quantify uncertainty in physical parameters for extrapolative prediction,
then there is no need to perform inference on the discrepancy term. With this
in mind, we introduce Gibbs posteriors as an alternative Bayesian method for
model calibration, which updates the prior with a loss function connecting the
data to the parameter. The target of inference is the physical parameter value
which minimizes the expected loss. We propose to tune the loss scale of the
Gibbs posterior to maintain nominal frequentist coverage under assumptions of
the form of model discrepancy, and present a bootstrap implementation for
approximating coverage rates. Our approach is highly modular, allowing an
analyst to easily encode a wide variety of such assumptions. Furthermore, we
provide a principled method of combining posteriors calculated from data
subsets. We apply our methods to data from an experiment measuring the material
properties of tantalum.
| stat |
Compound vectors of subordinators and their associated positive L\'evy
copulas | L\'evy copulas are an important tool which can be used to build dependent
L\'evy processes. In a classical setting, they have been used to model
financial applications. In a Bayesian framework they have been employed to
introduce dependent nonparametric priors which allow to model heterogeneous
data. This paper focuses on introducing a new class of L\'evy copulas based on
a class of subordinators recently appeared in the literature, called
\textit{Compound Random Measures}. The well-known Clayton L\'evy copula is a
special case of this new class. Furthermore, we provide some novel results
about the underlying vector of subordinators such as a series representation
and relevant moments. The article concludes with an application to a Danish
fire dataset.
| stat |
A Particle Filter for Stochastic Advection by Lie Transport (SALT): A
case study for the damped and forced incompressible 2D Euler equation | In this work, we combine a stochastic model reduction with a particle filter
augmented with tempering and jittering, and apply the combined algorithm to a
damped and forced incompressible 2D Euler dynamics defined on a simply
connected bounded domain.
We show that using the combined algorithm, we are able to assimilate data
from a reference system state (the ``truth") modelled by a highly resolved
numerical solution of the flow that has roughly $3.1\times10^6$ degrees of
freedom, into a stochastic system having two orders of magnitude less degrees
of freedom, which is able to approximate the true state reasonably accurately
for $5$ large scale eddy turnover times, using modest computational hardware.
The model reduction is performed through the introduction of a stochastic
advection by Lie transport (SALT) model as the signal on a coarser resolution.
The SALT approach was introduced as a general theory using a geometric
mechanics framework from Holm, Proc. Roy. Soc. A (2015). This work follows on
the numerical implementation for SALT presented by Cotter et al, SIAM
Multiscale Model. Sim. (2019) for the flow in consideration. The model
reduction is substantial: The reduced SALT model has $4.9\times 10^4$ degrees
of freedom.
Results from reliability tests on the assimilated system are also presented.
| stat |
Exact risk improvement of bandwidth selectors for kernel density
estimation with directional data | New bandwidth selectors for kernel density estimation with directional data
are presented in this work. These selectors are based on asymptotic and exact
error expressions for the kernel density estimator combined with mixtures of
von Mises distributions. The performance of the proposed selectors is
investigated in a simulation study and compared with other existing rules for a
large variety of directional scenarios, sample sizes and dimensions. The
selector based on the exact error expression turns out to have the best
behaviour of the studied selectors for almost all the situations. This selector
is illustrated with real data for the circular and spherical cases.
| stat |
Supervised Autoencoders Learn Robust Joint Factor Models of Neural
Activity | Factor models are routinely used for dimensionality reduction in modeling of
correlated, high-dimensional data. We are particularly motivated by
neuroscience applications collecting high-dimensional `predictors'
corresponding to brain activity in different regions along with behavioral
outcomes. Joint factor models for the predictors and outcomes are natural, but
maximum likelihood estimates of these models can struggle in practice when
there is model misspecification. We propose an alternative inference strategy
based on supervised autoencoders; rather than placing a probability
distribution on the latent factors, we define them as an unknown function of
the high-dimensional predictors. This mapping function, along with the
loadings, can be optimized to explain variance in brain activity while
simultaneously being predictive of behavior. In practice, the mapping function
can range in complexity from linear to more complex forms, such as splines or
neural networks, with the usual tradeoff between bias and variance. This
approach yields distinct solutions from a maximum likelihood inference
strategy, as we demonstrate by deriving analytic solutions for a linear
Gaussian factor model. Using synthetic data, we show that this function-based
approach is robust against multiple types of misspecification. We then apply
this technique to a neuroscience application resulting in substantial gains in
predicting behavioral tasks from electrophysiological measurements in multiple
factor models.
| stat |
Multi-fidelity Gaussian Process Bandit Optimisation | In many scientific and engineering applications, we are tasked with the
maximisation of an expensive to evaluate black box function $f$. Traditional
settings for this problem assume just the availability of this single function.
However, in many cases, cheap approximations to $f$ may be obtainable. For
example, the expensive real world behaviour of a robot can be approximated by a
cheap computer simulation. We can use these approximations to eliminate low
function value regions cheaply and use the expensive evaluations of $f$ in a
small but promising region and speedily identify the optimum. We formalise this
task as a \emph{multi-fidelity} bandit problem where the target function and
its approximations are sampled from a Gaussian process. We develop MF-GP-UCB, a
novel method based on upper confidence bound techniques. In our theoretical
analysis we demonstrate that it exhibits precisely the above behaviour, and
achieves better regret than strategies which ignore multi-fidelity information.
Empirically, MF-GP-UCB outperforms such naive strategies and other
multi-fidelity methods on several synthetic and real experiments.
| stat |
Non-parametric efficient causal mediation with intermediate confounders | Interventional effects for mediation analysis were proposed as a solution to
the lack of identifiability of natural (in)direct effects in the presence of a
mediator-outcome confounder affected by exposure. We present a theoretical and
computational study of the properties of the interventional (in)direct effect
estimands based on the efficient influence fucntion (EIF) in the non-parametric
statistical model. We use the EIF to develop two asymptotically optimal,
non-parametric estimators that leverage data-adaptive regression for estimation
of the nuisance parameters: a one-step estimator and a targeted minimum loss
estimator. A free and open source \texttt{R} package implementing our proposed
estimators is made available on GitHub. We further present results establishing
the conditions under which these estimators are consistent, multiply robust,
$n^{1/2}$-consistent and efficient. We illustrate the finite-sample performance
of the estimators and corroborate our theoretical results in a simulation
study. We also demonstrate the use of the estimators in our motivating
application to elucidate the mechanisms behind the unintended harmful effects
that a housing intervention had on adolescent girls' risk behavior.
| stat |
Replication or exploration? Sequential design for stochastic simulation
experiments | We investigate the merits of replication, and provide methods for optimal
design (including replicates), with the goal of obtaining globally accurate
emulation of noisy computer simulation experiments. We first show that
replication can be beneficial from both design and computational perspectives,
in the context of Gaussian process surrogate modeling. We then develop a
lookahead based sequential design scheme that can determine if a new run should
be at an existing input location (i.e., replicate) or at a new one (explore).
When paired with a newly developed heteroskedastic Gaussian process model, our
dynamic design scheme facilitates learning of signal and noise relationships
which can vary throughout the input space. We show that it does so efficiently,
on both computational and statistical grounds. In addition to illustrative
synthetic examples, we demonstrate performance on two challenging real-data
simulation experiments, from inventory management and epidemiology.
| stat |
Powering population health research: Considerations for plausible and
actionable effect sizes | Evidence for Action (E4A), a signature program of the Robert Wood Johnson
Foundation, funds investigator-initiated research on the impacts of social
programs and policies on population health and health inequities. Across
thousands of letters of intent and full proposals E4A has received since 2015,
one of the most common methodological challenges faced by applicants is
selecting realistic effect sizes to inform power and sample size calculations.
E4A prioritizes health studies that are both (1) adequately powered to detect
effect sizes that may reasonably be expected for the given intervention and (2)
likely to achieve intervention effects sizes that, if demonstrated, correspond
to actionable evidence for population health stakeholders. However, little
guidance exists to inform the selection of effect sizes for population health
research proposals. We draw on examples of five rigorously evaluated population
health interventions. These examples illustrate considerations for selecting
realistic and actionable effect sizes as inputs to power and sample size
calculations for research proposals to study population health interventions.
We show that plausible effects sizes for population health inteventions may be
smaller than commonly cited guidelines suggest. Effect sizes achieved with
population health interventions depend on the characteristics of the
intervention, the target population, and the outcomes studied. Population
health impact depends on the proportion of the population receiving the
intervention. When adequately powered, even studies of interventions with small
effect sizes can offer valuable evidence to inform population health if such
interventions can be implemented broadly. Demonstrating the effectiveness of
such interventions, however, requires large sample sizes.
| stat |
Multiscale Non-stationary Stochastic Bandits | Classic contextual bandit algorithms for linear models, such as LinUCB,
assume that the reward distribution for an arm is modeled by a stationary
linear regression. When the linear regression model is non-stationary over
time, the regret of LinUCB can scale linearly with time. In this paper, we
propose a novel multiscale changepoint detection method for the non-stationary
linear bandit problems, called Multiscale-LinUCB, which actively adapts to the
changing environment. We also provide theoretical analysis of regret bound for
Multiscale-LinUCB algorithm. Experimental results show that our proposed
Multiscale-LinUCB algorithm outperforms other state-of-the-art algorithms in
non-stationary contextual environments.
| stat |
Estimation of Partially Conditional Average Treatment Effect by Hybrid
Kernel-covariate Balancing | We study nonparametric estimation for the partially conditional average
treatment effect, defined as the treatment effect function over an interested
subset of confounders. We propose a hybrid kernel weighting estimator where the
weights aim to control the balancing error of any function of the confounders
from a reproducing kernel Hilbert space after kernel smoothing over the subset
of interested variables. In addition, we present an augmented version of our
estimator which can incorporate estimations of outcome mean functions. Based on
the representer theorem, gradient-based algorithms can be applied for solving
the corresponding infinite-dimensional optimization problem. Asymptotic
properties are studied without any smoothness assumptions for propensity score
function or the need of data splitting, relaxing certain existing stringent
assumptions. The numerical performance of the proposed estimator is
demonstrated by a simulation study and an application to the effect of a
mother's smoking on a baby's birth weight conditioned on the mother's age.
| stat |
Quantile Graphical Models: Bayesian Approaches | Graphical models are ubiquitous tools to describe the interdependence between
variables measured simultaneously such as large-scale gene or protein
expression data. Gaussian graphical models (GGMs) are well-established tools
for probabilistic exploration of dependence structures using precision matrices
and they are generated under a multivariate normal joint distribution. However,
they suffer from several shortcomings since they are based on Gaussian
distribution assumptions. In this article, we propose a Bayesian quantile based
approach for sparse estimation of graphs. We demonstrate that the resulting
graph estimation is robust to outliers and applicable under general
distributional assumptions. Furthermore, we develop efficient variational Bayes
approximations to scale the methods for large data sets. Our methods are
applied to a novel cancer proteomics data dataset wherein multiple proteomic
antibodies are simultaneously assessed on tumor samples using reverse-phase
protein arrays (RPPA) technology.
| stat |
Modelling provincial Covid-19 epidemic data in Italy using an adjusted
time-dependent SIRD model | In this paper we develop a predictive model for the spread of COVID-19
infection at a provincial (i.e. EU NUTS-3) level in Italy by using official
data from the Italian Ministry of Health integrated with data extracted from
daily official press conferences of regional authorities and from local
newspaper websites. This integration is mainly concerned with COVID-19 cause
specific death data which are not available at NUTS-3 level from open official
data data channels. An adjusted time-dependent SIRD model is used to predict
the behavior of the epidemic, specifically the number of susceptible, infected,
deceased and recovered people. Predictive model performance is evaluated using
comparison with real data.
| stat |
Radioactive data: tracing through training | We want to detect whether a particular image dataset has been used to train a
model. We propose a new technique, \emph{radioactive data}, that makes
imperceptible changes to this dataset such that any model trained on it will
bear an identifiable mark. The mark is robust to strong variations such as
different architectures or optimization methods. Given a trained model, our
technique detects the use of radioactive data and provides a level of
confidence (p-value). Our experiments on large-scale benchmarks (Imagenet),
using standard architectures (Resnet-18, VGG-16, Densenet-121) and training
procedures, show that we can detect usage of radioactive data with high
confidence (p<10^-4) even when only 1% of the data used to trained our model is
radioactive. Our method is robust to data augmentation and the stochasticity of
deep network optimization. As a result, it offers a much higher signal-to-noise
ratio than data poisoning and backdoor methods.
| stat |
An Approach to Statistical Process Control that is New, Nonparametric,
Simple, and Powerful | To maintain the desired quality of a product or service it is necessary to
monitor the process that results in the product or service. This monitoring
method is called Statistical Process Management, or Statistical Process
Control. It is in widespread usage in industry. Extensive statistical
methodology has been developed to make it possible to detect when a process
goes out of control while allowing for natural variability that occurs when the
process is in control. This paper introduces nonparametric methods for
monitoring data, whether it is univariate or multivariate, and whether the
interest is in detecting a change of location or scale or both. These methods,
based on sequential normal scores, are much simpler than the most popular
nonparametric methods currently in use and have good power for detecting out of
control observations. Sixteen new statistical tests are introduced for the
first time in this paper, with 17 examples, 33 tables, and 48 figures to
complete the instructions for their application.
| stat |
The Illusion of the Illusion of Sparsity: An exercise in prior
sensitivity | The emergence of Big Data raises the question of how to model economic
relations when there is a large number of possible explanatory variables. We
revisit the issue by comparing the possibility of using dense or sparse models
in a Bayesian approach, allowing for variable selection and shrinkage. More
specifically, we discuss the results reached by Giannone, Lenza, and Primiceri
(2020) through a "Spike-and-Slab" prior, which suggest an "illusion of
sparsity" in economic data, as no clear patterns of sparsity could be detected.
We make a further revision of the posterior distributions of the model, and
propose three experiments to evaluate the robustness of the adopted prior
distribution. We find that the pattern of sparsity is sensitive to the prior
distribution of the regression coefficients, and present evidence that the
model indirectly induces variable selection and shrinkage, which suggests that
the "illusion of sparsity" could be, itself, an illusion. Code is available on
github.com/bfava/IllusionOfIllusion.
| stat |
The economic value of additional airport departure capacity | This article presents a model for the economic value of extra capacity at an
airport. The model is based on a series of functional relationships linking the
benefits of extra capacity and the associated costs. It takes into account the
cost of delay for airlines and its indirect consequences on the airport,
through the loss or gain of aeronautical and non-aeronautical revenues. The
model is highly data-driven and to this end a number of data sources have been
used. In particular, special care has been used to take into account the full
distribution of delay at the airports rather than its average only. The results
with the simple version of the model show the existence of a unique maximum for
the operating profit of the airport in terms of capacity. The position of this
maximum is clearly dependent on the airport and also has an interesting
behaviour with the average number of passenger per aircraft at the airport and
the predictability of the flight departure times. In addition, we also show
that there exists an important trade-off between an increased predictability
and the punctuality at the airport. Finally, it is shown that a more complex
behavioural model for passengers can introduce several local maxima in the
airport profit and thus drive the airport towards suboptimal decisions.
| stat |
Q-FIT: The Quantifiable Feature Importance Technique for Explainable
Machine Learning | We introduce a novel framework to quantify the importance of each input
feature for model explainability. A user of our framework can choose between
two modes: (a) global explanation: providing feature importance globally across
all the data points; and (b) local explanation: providing feature importance
locally for each individual data point. The core idea of our method comes from
utilizing the Dirichlet distribution to define a distribution over the
importance of input features. This particular distribution is useful in ranking
the importance of the input features as a sample from this distribution is a
probability vector (i.e., the vector components sum to 1), Thus, the ranking
uncovered by our framework which provides a \textit{quantifiable explanation}
of how significant each input feature is to a model's output. This quantifiable
explainability differentiates our method from existing feature-selection
methods, which simply determine whether a feature is relevant or not.
Furthermore, a distribution over the explanation allows to define a closed-form
divergence to measure the similarity between learned feature importance under
different models. We use this divergence to study how the feature importance
trade-offs with essential notions in modern machine learning, such as privacy
and fairness. We show the effectiveness of our method on a variety of synthetic
and real datasets, taking into account both tabular and image datasets.
| 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.