Datasets:

SouthernCrossAI
/

Arxiv

Dataset card Data Studio Files Files and versions Community

Split (1)

train · 132k rows

Filename stringlengths 22 64	Paragraph stringlengths 8 5.57k
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	systems. Fundamentally, however, interbank money markets are ﬂow networks, in which links between the nodes correspond to monetary ﬂows. The dynamics of such ﬂows has not been examined in depth in previous studies, which mostly viewed interbank networks as static or slowly varying. But the underlying ﬂows are highly dynamic and complex. Moreover, monetary ﬂows are inhomogeneous; loan ﬂows are fundamentally diﬀerent from the ﬂows of other payments. Payments by the banks’ customers and the banks themselves cause imbalances in the exchange settlement accounts of the banks. For some banks, the incoming ﬂows exceed the outgoing ﬂows on any given day; for other banks, the reverse is true. Banks with excess reserves lend them in the overnight money market to banks with depleted reserves. This creates interesting dynamics: payment ﬂows cause imbalances, which in turn drive compensating ﬂows of loans. Understanding this dynamic relationship is needed for advancing our ability to model interbank markets eﬀectively.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	In this paper, our objective is to deﬁne empirically the dynamics of interbank monetary ﬂows. Unlike most studies cited above, we aim to uncover the fundamental causal relationship between the ﬂows of overnight loans and other payments. We choose to specialise in the Australian interbank market, where we have privileged access to a high-quality dataset provided by the Reserve Bank of Australia (RBA). Our dataset consists of transactions settled in the period from 19 to 23 February 2007 in the Australian interbank market. We separate overnight loans and other payments (which we call nonloans) using a standard matching procedure. The loan and nonloan transactions settled on a given day form the ﬂow networks, which are the main target of our statistical analysis. We compare the topology and variation of the loan and nonloan networks and reveal the causal mechanism that ties them together. We investigate the dynamical stability of the system by testing how individual ﬂows vary from day to day. Basic network properties such as the degree distribution and assortativity are examined as well.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	High-value transactions between Australian banks are settled via the Reserve Bank Information and Transfer System (RITS) operated by the RBA since 1998 on an RTGS basis [20]. The transactions are settled continuously throughout the day by crediting and debiting the exchange settlement accounts held by the RBA on behalf of the participating banks. The banks’ exchange settlement accounts at the RBA are continuously monitored to ensure liquidity, with provisions for intra-day borrowing via the intra-day liquidity facility provided to the qualifying banks by the RBA. This obviates the need for a monthly reserve cycle of the sort maintained by Italian banks as discussed in [13]. The RITS is used as a feeder system for transactions originating from SWIFT1 and Austraclear for executing foreign exchange and securities transactions respectively. The member banks can also enter transactions directly into RITS. The switch to real-time settlement in 1998 was an important reform which protects the payment system against systemic risk, since transactions can only be settled if the paying banks possess suﬃcient funds in their exchange settlement accounts. At present, about 3.2 × 104 transactions are settled per day, with total value around A$168 billion.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The data comprise all interbank transfers processed on an RTGS basis by the RBA during the week of 19 February 2007. During this period, 55 banks participated in the RITS including the RBA. The dataset includes transfers between the banks and the RBA, such as RBA’s intra-day repurchase agreements and money market operations. The real bank names are obfuscated (replaced with labels from A to BP) for privacy reasons, but the obfuscated labels are consistent over the week. The transactions are grouped into separate days, but the time stamp of each transaction is removed.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	During the week in question, around 2.5 × 104 transactions were settled per day, with the total value of all transactions rising above A$2 × 1011 on Tuesday and Thursday. The number of transactions (volume2) and the total value (the combined dollar amount of all transactions) for each day are given in Table 1.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Table 1: The number of transactions (volume) and their total value (in units of A$109) for each day.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Table 2: Mean hui, variance σ2 Figure 1 (u = log10 v, where v is value).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 1 shows the distribution of transaction values on a logarithmic scale. Local peaks in the distribution correspond to round values. The most pronounced peak occurs at A$106.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	In terms of the number of transactions, the distribution consists of two approximately log-normal com- ponents, with lower-value transactions being slightly more numerous. The standard entropy maximisation algorithm for a Gaussian mixture model with two components [21] produces a satisfactory ﬁt with the pa- rameters indicated in Table 2. The lower- and higher-value components are typically centred around A$104 and A$106 respectively. The high-value component is small on Monday (19-02-2007) but increases notice- ably on subsequent days, while the low-value component diminishes. By value, however, the distribution is clearly dominated by transactions above A$106, with the highest contribution from around A$2 × 108.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The target interest rate of the RBA during the week of our sample was rt = 6.25% per annum. If the target rate is known, it is easy to extract the overnight loans from the data by identifying reversing transactions on consecutive days. A hypothetical interest rate can be computed for each reversing transaction and compared with the target rate. For instance, suppose a transaction of value v1 from bank A to bank B on day 1 reverses with value v2, from bank B to bank A, on day 2. These transactions are candidates for the ﬁrst and second legs of an overnight loan from A to B. The hypothetical interest rate for this pair of transactions is given by rh = 100% × 365 × (v2 − v1)/v1; note that the quoted target rate is per annum. Since large banks participate in many reversing transactions that can qualify as loans, we consider all possible hypothetical pairs and prefer the one that gives rh closest to the target rate. The algorithm for loan extraction is applied from Monday to Thursday; loans issued on Friday cannot be processed since the next day is not available. A similar procedure was pioneered by Furﬁne [22]; see also [23].
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 1: The distribution of transaction values v (in Australian dollars) on a logarithmic scale, with bin size ∆ log10 v = 0.1; the vertical axis is the number of transactions per bin. Components of the Gaussian mixture model are indicated by the dashed curves; the solid curve is the sum of the two components. The dotted histogram shows the relative contribution of transactions at diﬀerent values to the total value (to compute the dotted histogram we multiply the number of transactions in a bin by their value).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	from the target rate; and the lower-value transactions (below A$104), even if they are real loans, contribute negligibly to the total value.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	We identify 897 overnight loans over the four days. A daily breakdown is given in Table 3. Here and below, we refer to the ﬁrst leg of the overnight loans as simply loans and to all other transactions as nonloans. The loans constitute less than 1% of all transactions by number and up to 9% by value (cf. Tables 1 and 3). The distribution of loan values and interest rates is shown in Figures 3a and 3b. The interest rate distribution peaks at the target rate 6.25%. The mean rate is within one basis point (0.01%) of the target rate, while the standard deviation is about 0.07%. The average interest rate increases slightly with increasing value of the loan; a least-squares ﬁt yields rh = 6.248 + 0.010 log10(v/A$106).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The same technique can be used to extract two-day and longer-term loans (up to four-day loans for our sample of ﬁve consecutive days). Using the same selection criteria as for the overnight loans, our algorithm detects 27, 67, and 24 two-day loans, with total values A$1.3, A$2.2, and A$1.4 billion, on Monday, Tuesday, and Wednesday, respectively. The total value of the two-day loans is 1.5%, 1.0%, and 0.9% of the total transaction values on these days respectively.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 2: Hypothetical interest rate rh versus value of the ﬁrst leg of the transaction pairs detected by our algorithm, with no restrictions on value or interest rate. The dotted rectangle contains the transactions that we identify as overnight loans. The least-squares ﬁt is shown with a solid red line.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Table 3: Statistics of the overnight loans identiﬁed by our algorithm: the number of loans (volume), the total value of the ﬁrst leg of the loans (in units of A$109), and the fraction of the total value of the loans (ﬁrst legs only) with respect to the total value of all transactions on a given date.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	also an unusually large number of A$106 and A$400 transactions from W to T on Monday. Note that the daily imbalance for each bank is mostly determined by the highest value transactions; large discrepancies between incoming and outgoing transactions at lower values are less relevant.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The distribution for BA is clearly bimodal; it contains an unusually high proportion of transactions greater than A$106. Moreover, below A$106, incoming transactions typically outnumber outgoing ones by a large amount. BA is also involved in many high value transactions that reverse on the same day. These transactions probably correspond to the central bank’s repurchase agreements, which facilitate intra-day liquidity of the banks [24].
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The banks shown in Figure 4 are also the largest in term of the number of transactions, with the exception of BA. The rank order by the number of transactions matches that by value. For D, which is the largest, the number of nonloan transactions reaches 48043 over the week. By the number of transactions, the order of the top twelve banks is D, BP, AV, T, W, AH, AF, U, AP, BI, BA, P. By value, the order is D, BP, AV, BA, T, W, BG, U, A, AH, AB, BM. The situation is similar when considering the overnight loans. By value, AV, D, BP, and T dominate. For these four banks, weekly total loans range from A$11.5 to A$18 billion and number from 254 to 399. For the other banks the total loan value is less than A$3 billion.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 3: (a) The distribution of loan values v on a logarithmic scale. The vertical axis is the number of loans per bin for bin size ∆ log10 v = 0.25. The dotted line is the same distribution multiplied by the value corresponding to each bin (in arbitrary units). The date of the ﬁrst leg of the loans is indicated. (b) The distribution of loan interest rates rh. The vertical axis is the number of loans per bin for bin size ∆rh = 0.01. The date of the ﬁrst leg of the loans is indicated. The mean and standard deviation are 6.25% and 0.08% on Monday (19-02-2007), and 6.26% and 0.07% on the other days.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	four and the rest of the banks in terms of cash and liquid assets at the time was less clear, with Macquarie Bank Limited in third position with A$8 billion. According to APRA, cash and liquid assets of the big four and Macquarie Bank Limited accounted for 56% of the total.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 4: The distribution of nonloan transaction values of the six largest banks for Monday through Thursday (from left to right); the banks are selected by the combined value of incoming and outgoing transactions over the entire week. Black and red histograms correspond to incoming (bank is the destination) and outgoing (bank is the source) transactions; red histograms are ﬁlled in to improve visibility. The banks’ anonymous labels, the combined daily value of the incoming and outgoing transactions, and the daily imbalance (incoming minus outgoing) are quoted at the top left of each panel (in units of A$109). The horizontal axis is the logarithm of value in A$.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	of a corporate customer are depleted, this customer may borrow overnight to replenish the funds. In this case, the overnight loan is initiated by an account holder, who generally has no knowledge of the bank’s net position. Nevertheless, the actions of this account holder in acquiring a loan reduce the bank’s imbalance, provided that the customer deposits the loan in an account with the same bank.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The loan and nonloan imbalances for the six largest banks are given in Table 4. The data generally comply with our assumption that the overnight loans compensate the daily imbalances of the nonloan transactions. The most obvious exception is for BA on Thursday (22-02-2007), where a large negative nonloan imbalance is accompanied by a sizable loan imbalance that is also negative. Taking all the banks together, there is a strong anti-correlation between loan and nonloan imbalances on most days. We see this clearly in Figure 5. The Pearson correlation coeﬃcients for Monday through Thursday are −0.93, −0.88, −0.95, −0.36. It is striking to observe that many points fall close to the perfect anti-correlation line. The anti-correlation is weaker on Thursday (crosses in Figure 5), mostly due to BA and AV.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	A correlation also exists between the absolute values of loan imbalances and the nonloan total values (incoming plus outgoing nonloan transactions); the Pearson coeﬃcients are 0.74, 0.75, 0.66, 0.77 for Monday through Thursday. This conﬁrms the intuitive expectation that larger banks tolerate larger loan imbalances.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 5: Left: loan imbalance ∆l vs nonloan imbalance ∆v for individual banks and days of the week (in units of A$109). Right: the absolute value of loan imbalance \|∆l\| vs nonloan total value (incoming plus outgoing transactions) for individual banks and days of the week. Thursday data are marked with crosses.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Table 4: Loan and nonloan imbalances for the six largest banks (in units of A$109).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	where {vk(i, j)}k is a list of values of individual nonloan transaction from i to j on the day. For example, all nonloan transactions from D to AV on Monday form a nonloan ﬂow from D to AV on that day. The nonloan transactions in the opposite direction, from AV to D, form another ﬂow. A ﬂow has zero value if the number of transactions is zero. Typically, for any two large banks there are two nonloan ﬂows between them. The loan ﬂows are computed in a similar fashion.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	There are 55 banks in the network, resulting in Nﬂow = 2970 possible ﬂows. The actual number of ﬂows is much smaller. The typical number of nonloan ﬂows is ∼ 800 on each day (the actual numbers are 804, 791, 784, 797). Even though the number of nonloan ﬂows does not change signiﬁcantly from day to day, we ﬁnd that only about 80% of these ﬂows persist for two days or more. The other 20% are replaced by diﬀerent ﬂows, i.e. with a diﬀerent source and/or destination, on the following day. Structurally speaking, the network of nonloan ﬂows changes by 20% from day to day. However, persistent ﬂows carry more than 96% of the total value.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 6: Nonloan ﬂow value pairs on one day (horizontal axis) and the next (vertical axis). Only ﬂows present on both days are considered. Flows that do not change lie on the diagonal (red dotted line). The solid line is the weighted orthogonal least squares ﬁt to the scatter diagram; the weights have been deﬁned to emphasize points corresponding to large ﬂows.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Tuesday, Tuesday and Wednesday, and Wednesday and Thursday. If the ﬂow values were the same, the points in Figure 6 would lie on the diagonals. We observe that the values of some ﬂows vary signiﬁcantly, especially when comparing Monday and Tuesday. Moreover, there is a notable systematic increase in value of the ﬂows from Monday to Tuesday by a factor of several, which is not observed on the other days. For each pair of days shown in Figure 6, we compute the Pearson correlation coeﬃcient, which gives 0.53 for Monday and Tuesday, 0.70 for Tuesday and Wednesday, and 0.68 for Wednesday and Thursday.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	To characterize the diﬀerence between the ﬂows on diﬀerent days more precisely, we compute the Eu- clidean distance between normalised ﬂows on consecutive days. We reorder the adjacency matrix {vﬂow(i, j)}ij of the ﬂow network on day d as an Nﬂow-dimensional vector vd representing a list of all ﬂows on day d (d = 1, 2, . . . , 5). For each pair of consecutive days we compute the Euclidean distance between normalized vectors vd/\|vd\| and vd+1/\|vd+1\|, which gives 0.62, 0.50, 0.50 for all ﬂows and 0.61, 0.49, 0.49 for persis- tent ﬂows (the latter are computed by setting non-persistent ﬂows to zero on both days). Since the ﬂow vectors are normalized, these quantities measure random ﬂow discrepancies while systematic deviation such as between the ﬂows on Monday and Tuesday are ignored. For two vectors of random values uniformly distributed in interval (0, 1), the expected Euclidean distance is 0.71 and the standard deviation is 0.02 for the estimated number of persistent nonloan ﬂows of 640. So the observed variability of the nonloan ﬂows is smaller than what one might expect if the ﬂow values were random.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Variability of the loan ﬂows is equally strong. The number of loan ﬂows varies from 69 to 83 (actual numbers are 69, 75, 77, 83). Only about 50% of these ﬂows are common for any two consecutive days. Moreover, persistent ﬂows carry only about 65% of the total value of the loan ﬂows on any given day, cf. 80% of nonloan ﬂows. For persistent loan ﬂows, the Pearson correlation coeﬃcients are 0.63, 0.90, and 0.76 for the consecutive pairs of days starting with Monday and Tuesday. The correlation is generally similar to that of the nonloan ﬂows, with the notable exception of the loan ﬂows on Tuesday and Wednesday, when the sub-network of persistent loan ﬂows appears to be more stable.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The Euclidean distances between the normalized loan ﬂows for each pair of consecutive days are 0.85, 0.68, 0.73 for all ﬂows and 0.63, 0.44, and 0.44 for persistent ﬂows. For two vectors of random values uniformly distributed in interval (0, 1), the expected Euclidean distance is 0.7 and the standard deviation is 0.1 for the estimated number of persistent loan ﬂows of 40. So the observed variability of the persistent loan ﬂows is much smaller than what one might expect if the ﬂow values were random.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Some loan ﬂows do not have corresponding nonloan ﬂows between the same nodes on the same day. These ﬂows carry about 14% of loan value on Monday, and about 7% on Tuesday through Thursday. Nonloan ﬂows that have corresponding loan ﬂows account for 35% to 48% of all nonloan ﬂows by value, even though the number of these ﬂows is less than 10% of the total.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	To improve the statistics, we aggregate the ﬂows on all four days. Figure 8 shows nonloan and cor- responding loan ﬂow values. We fail to ﬁnd any signiﬁcant correlation between loan and nonloan ﬂows (Pearson coeﬃcient is 0.3). The correlation improves if we restrict the loan ﬂows to those consisting of three transactions or more; such ﬂows mostly correspond to large persistent ﬂows. In this case the Pearson coeﬃcient increases to 0.6; banks that sustain large nonloan ﬂows can also sustain large loan ﬂows, even though the loan ﬂows on average are an order of magnitude lower than the corresponding nonloan ﬂows. The lack of correlation when all loans are aggregated is due to the presence of many large loans that are not accompanied by large nonloan transactions, and vice versa.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	For instance, if the ﬂow from D to AV is larger than the ﬂow in the opposite direction, then the net ﬂow is from D to AV.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The distributions of net loan and nonloan ﬂow values are presented in Figure 9. The parameters of the associated Gaussian mixture models are quoted in Table 5. The distribution of net nonloan ﬂow values has the same general features as the distribution of the individual transactions. However, unlike individual transactions, net ﬂow values below A$104 are rare; net ﬂows around A$108 are more prominent.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 8: Loan ﬂow values versus nonloan ﬂow values combined over four days. Triangles correspond to loan ﬂows with three or more transactions per ﬂow. The solid line is the orthogonal least squares ﬁt to the scatter diagram; the weighting is the same as in Figure 6.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	where Nnet(n) is the number of net nonloan ﬂows that consist of n transactions (n ranges from 1 to more than 1000). This is consistent with the ﬁndings for Fedwire reported in [15] (see right panel of Fig. 14 in [15]).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	There are roughly 60 net loan ﬂows each day. As many as 40 consist of only one transaction. On the other hand, a single net loan ﬂow between two large banks may comprise more than 30 individual loans. The distribution of the number of transactions per net loan ﬂow is diﬃcult to infer due to poor statistics, but it is consistent with a power law with a steeper exponent, −1.4 ± 0.2, than that of the nonloan distribution. There are no net loan ﬂows below A$105 or above A$109. Comparing net loan and nonloan ﬂows, it is obvious that net loan ﬂows cannot compensate each and every net nonloan ﬂow. Not only are there fewer net loan ﬂows than nonloan ﬂows, but the total value of the former is much less than the total value of the latter.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 9: The distribution of values of net nonloan ﬂows (black histogram) on a logarithmic scale with bin size ∆ log10 v = 0.1. The components of the Gaussian mixture model are indicated with the dashed curves; the solid curve is the sum of the two components. Net loan ﬂows are overplotted in red. The vertical axis counts the number of net ﬂows per bin.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Net loan and net nonloan ﬂows are not correlated; the correlation coeﬃcient is 0.3. Restricting net loan ﬂows to those that have three transactions or more does not improve the correlation. If a net loan ﬂow between two banks was triggered to a signiﬁcant degree by the magnitude and the direction of net nonloan ﬂow between these bank, one expects a correlation between net loan and nonloan ﬂows. Our examination shows that in this respect loan ﬂows are decoupled from nonloan ﬂows. The connection between them is indirect. Namely, nonloan ﬂows cause an imbalance in the account of each bank, which is subsequently compensated by loan ﬂows, which are largely unrelated to the nonloan ﬂows that caused the imbalance.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	We deﬁne the in-degree of node i as the number of net ﬂows that terminate at i, i.e. the number of net ﬂows with destination i, and the out-degree as the number of net ﬂows that originate from i, i.e. the number of net ﬂows with source i. The degree distribution of the nonloan networks is shown in Figure 10a. Node BA has the highest in-degree of 37 on Monday, but on the other days it drops to 15 on average, while the out-degree is 11.75 on average for this node. The highest in-degrees are usually found among the four largest banks (D, BP, AV, T); the only exception is Monday, when AF’s in-degree of 22 is greater than AV’s 21, and BA has the highest in-degree. The highest out-degrees are usually achieved by D, BP, AV, T, W, and AH; the exceptions are Monday, when D’s out-degree of 17 is less than AR’s and AP’s 18, and Thursday, when AV’s out-degree of 16 is less than P’s 18.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	It is diﬃcult to infer the shape of the degree distribution for individual days due to poor statistics. The two-sample Kolmogorov-Smirnov (KS) test does not distinguish between the distributions on diﬀerent days at the 5% signiﬁcance level. With this in mind, we combine the in- and out-degree data for all four days and graph the resulting distributions in Figure 10b. We ﬁnd that a power law distribution does not provides a good ﬁt for either in- or out-degrees. Visually, the distribution is closer to an exponential. However, the exponential distribution is rejected by the Anderson-Darling test.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 10: (a) Degree distribution of the net nonloan ﬂow networks (for convenience, in-degrees are positive and out-degrees are negative). The total value of the net ﬂows corresponding to the speciﬁc degrees is shown with red dots (the log of value in A$109 is indicated on the right vertical axis). (b) Degree distribution of the net nonloan ﬂows when the degree data for all four days are aggregated (in-degrees are circles; out-degrees are triangles).
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	with lower degrees are numerous, but the ﬂows they sustain are typically smaller than those carried by a few high-degree nodes. In particular, for the nonloan ﬂows, nodes with in-degree d ≤ 10 are numerous, ranging from 35 to 37, but their outgoing net ﬂows carry about 20% of the value on average. On the other hand, nodes with d ≥ 17 are rare, but their ﬂows carry 50% of the value. The same eﬀect is observed for the out-degrees.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The degree distribution of the network of net loan ﬂows is shown in Figure 11a (we ignore the nodes that have zero in- and out- degrees over four days). Similarly to nonloan ﬂows, the KS test does not distinguish between the distributions on diﬀerent days at the 5% signiﬁcance level. The combined distribution is shown in Figure 11b.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 11: (a) Same as Figure 10a, but for the net loan ﬂow networks. (b) Same as Figure 10b, but for the net loan ﬂows.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	the in-assortativity is −0.16, −0.26, −0.18, −0.19 and the out-assortativity is −0.03, −0.10, 0.02, −0.20 for the same sequence of days. In biological networks, the tendency of out-assortativity to be more assortative than in-assortativity has been noted in [25].
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Given the source and destination of each net ﬂow, we can construct a network representation of the net ﬂows. An example of the network of net nonloan ﬂows is shown in Figure 12. The size of the nodes and the thickness of the edges are proportional to the net imbalances and net ﬂow values respectively (on a logarithmic scale). We use the Fruchterman-Reingold algorithm to position the nodes [26]; the most connected nodes are placed in the centre, and the least connected nodes are moved to the periphery. The core of the network is dominated by the four banks with the largest total value and the largest number of transactions: D, BP, AV, and T. The other big banks, such as AF, AH, and W, also sit near the core. It is interesting to note the presence of several poorly connected nodes (Q, V, BF, and especially X) that participate in large incoming and outgoing ﬂows, which produce only negligible imbalances in the banks themselves.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The sub-network consisting of D, BP, AV, BA, T, W, U, A, AH, AF, AP, and P is fully connected on all ﬁve days, i.e. every node is connected to every other node. The sub-network of D, AV, and BP is fully connected, even if we restrict the net ﬂows to values above A$108.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 12: Network of net nonloan ﬂows on Tuesday, 20-02-2007. White (grey) nodes represent negative (positive) imbalances. The bank labels are indicated for each node. The size of the nodes and the thickness of the edges are proportional to the logarithm of value of the imbalances and the net ﬂows respectively.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	plots, we employ the weighted Fruchterman-Reingold algorithm, which positions the nodes with large ﬂows between them close to each other. The imbalances shown in Figure 13b are the same as those of the full network in Figure 12. The daily networks of net loan ﬂows for the same nodes are shown in Figures 14a–14d. We observe that the largest ﬂows on Monday (19-02-2007) were signiﬁcantly lower than the ﬂows on the subsequent days. The largest nodes (D, BP, AV, T, W) are always placed close to the center of the network, because they participate in the largest ﬂows. The topology of the ﬂows is complex and diﬃcult to disentangle, even if one concentrates on the largest ﬂows (above A$5 × 108). For instance, on Monday, probably the simplest day, the ﬂow of nonloans is generally from BG to “others” to D to BP. There are also sizable ﬂows from T to AV and from AV to “others” and BP. However, lower value ﬂows (below A$5 × 108) cannot be neglected completely because they are numerous and may contribute signiﬁcantly to the imbalance of a given node.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Nodes D, T, BP, AV, and W form a complete sub-network of net loan ﬂows on Monday, Tuesday, and Wednesday. This sub-network is almost complete on Thursday too, except for the missing link between BP and W. The appearance of the net loan network is diﬀerent from that of the nonloan network, since the same nodes participate in only a few loan ﬂows. Therefore, the position of a node in the network image is strongly inﬂuenced by the number of connections of that node. Some of the poorly connected nodes are placed at the periphery despite the fact that they possess large ﬂows. The four largest nodes (D, T, BP, AV) are always positioned at the center of the network.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The net nonloan ﬂow network is extremely volatile in terms of ﬂow value and direction. For example, a A$109 ﬂow from D to BP on Monday transforms into a A$3.2 × 109 ﬂow in the same direction on Tuesday, only to be replaced by a A$6.3 × 108 ﬂow in the opposite direction on Wednesday, which diminishes further to A$2.5 × 109 on Thursday. Nodes T and BP display a similar pattern of reversing ﬂows between Tuesday and Wednesday. On the other hand, the net ﬂow between T and AV maintains the same direction, but the ﬂow value is strongly ﬂuctuating. In particular, a moderate A$4.8 × 108 ﬂow on Monday rises to A$1.9 × 109 on Tuesday, then falls sharply to A$2 × 108 on Wednesday and again rises to A$2.2 × 109 on Thursday.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Considering any three nodes, we observe that circular and transitive ﬂows are present on most days, the latter being more common. The most obvious example is a circular ﬂow between D, T, and BP on Thursday and a transitive ﬂow involving BG, T, and AV on the same day. The circular ﬂows are unstable in the sense that they do not persist over two days or more.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The net loan ﬂow network exhibits similar characteristics. Few net loan ﬂows persist over the four days. For example, the ﬂow from AV to T has the same direction and is similar in value on all four days. Circular loan ﬂows are also present, as the ﬂow between AV, T, and BP on Thursday demonstrates.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	In this paper, we study the properties of the transactional ﬂows between Australian banks participating in RITS. The value distribution of transactions is approximated well by a mixture of two log-normal compo- nents, possibly reﬂecting the diﬀerent nature of transactions originating from SWIFT and Austraclear. For the largest banks, the value distributions of incoming and outgoing transactions are similar. On the other hand, the central bank displays a high asymmetry between the incoming and outgoing transactions, with the former clearly dominating the latter for transactions below A$106.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Using a matching algorithm for reversing transactions, we successfully separate transactions into loans and nonloans. For overnight loans, we estimate the identiﬁcation rate at 98%. The mean derived interest rate is within 0.01% of the central banks’ target rate of 6.25%, while the standard deviation is about 0.07%. We ﬁnd a strong anti-correlation between loan and nonloan imbalances (Pearson coeﬃcient is about 0.9 on most days). A likely explanation is that nonloan ﬂows create surpluses in some banks. The banks lend the surplus to banks in deﬁcit, creating loan ﬂows that counteract the imbalances due to the nonloan ﬂows. Hence, loan and nonloan imbalances of individual banks are roughly equal in value and opposite in sign on any given day.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 13: Networks of daily net nonloan ﬂows for D, AV, BP, T, W, BA, AH, AF, U, AP, P, A, BG. All the other nodes and the ﬂows to and from them are combined in a single new node called “others”. The size of the nodes and the thickness of the edges are proportional to the logarithm of value of the imbalances and the net ﬂows respectively. The value of the ﬂows and the imbalances can be gauged by referencing a network shown in the middle, where the values of the ﬂows are indicated in units of A$1 billion.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Figure 14: Networks of daily net loan ﬂows. The same nodes as in Figures 13a–13d are used. The scale of the loan ﬂows, the imbalances, and the positions of the nodes are the same as those used for the nonloan ﬂows in Figures 13a–13d to simplify visual comparison.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	The ﬂow networks are structurally variable, with 20% of nonloan ﬂows and 50% of loan ﬂows replaced every day. Values of persistent ﬂows, which maintain the same source and destination over at least two consecutive days, vary signiﬁcantly from day to day. Some ﬂow values change by several orders of magnitude. Persistent ﬂows increase in value several-fold between Monday and Tuesday. Individual ﬂow values can change by several orders of magnitude on the following day. Overall, there is a reasonable correlation between the ﬂow values on consecutive days (Pearson coeﬃcient is 0.65 for nonloans and 0.76 for loans on average). We also ﬁnd that larger banks tend to sustain larger loan ﬂows, in accord with the intuitive expectations. However, there is no correlation between loan and nonloan ﬂows.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	We examine visually the topology of the net loan and nonloan ﬂow networks. The centre of both networks is dominated by the big four banks. Twelve banks form a complete nonloan sub-network, in which each bank is connected to every other bank in the sub-network. The three largest banks form a complete sub- network even if the net ﬂows are restricted to values above A$108. Our examination reveals that the network topology of net ﬂows is complicated, with even the largest ﬂows varying greatly in value and direction on diﬀerent days.
Processed_Loan_and_nonloan_flows_in_the_Australian_interbank.txt	Our ﬁndings suggest a number of avenues for future research on interbank networks. Firstly, the relation- ships we uncovered can be used to constrain analytical models and numerical simulations of interbank ﬂows in ﬁnancial networks. In particular, our explanation of the link between the loan and nonloan imbalances needs to be tested in numerical simulations. Secondly, it is necessary to analyse interbank markets in other countries to establish what elements of our results are signatures of general dynamics and what aspects are speciﬁc to the epoch and location of this study. Even when high quality data are available, most previous studies concentrate on analysing static topological properties of the networks or their slow change over time. The internal dynamics of monetary ﬂows in interbank networks has been largely ignored. Importantly, one must ask whether the strong anti-correlation between loan and nonloan imbalances is characteristic of RTGS systems whose institutional setup resembles the Australian one or whether it is a general feature. For instance, in Italy a reserve requirement of 2% must be observed on the 23rd of each month, which may encourage strong deviations between loan and nonloan imbalances on the other days.
Processed_DASH:_A_C++_PGAS_Library_for_Distributed_Data_Stru.txt	Abstract—We present DASH, a C++ template library that offers distributed data structures and parallel algorithms and im- plements a compiler-free PGAS (partitioned global address space) approach. DASH offers many productivity and performance features such as global-view data structures, efﬁcient support for the owner-computes model, ﬂexible multidimensional data distribution schemes and inter-operability with STL (standard template library) algorithms. DASH also features a ﬂexible representation of the parallel target machine and allows the exploitation of several hierarchically organized levels of locality through a concept of Teams. We evaluate DASH on a number of benchmark applications and we port a scientiﬁc proxy application using the MPI two-sided model to DASH. We ﬁnd that DASH offers excellent productivity and performance and demonstrate scalability up to 9800 cores.
Processed_DASH:_A_C++_PGAS_Library_for_Distributed_Data_Stru.txt	The PGAS (Partitioned Global Address Space) model is a promising approach for programming large-scale systems [1], [2], [3]. When dealing with unpredictable and irregular com- munication patterns, such as those arising from graph analytics and data-intensive applications, the PGAS approach is often better suited and more convenient than two-sided message passing [4]. The PGAS model can be seen as an extension of threading-based shared memory programming to distributed memory systems, most often employing one-sided commu- nication primitives based on RDMA (remote direct memory access) mechanisms [5]. Since one-sided communication de- couples data movement from process synchronization, PGAS models are also potentially more efﬁcient than classical two- sided message passing approaches [6].
Processed_DASH:_A_C++_PGAS_Library_for_Distributed_Data_Stru.txt	However, PGAS approaches have so far found only limited acceptance and adoption in the HPC community [7]. One reason for this lack of widespread usage is that for PGAS languages, such as UPC [8], Titanium [9], and Chapel [10], adopters are usually required to port the whole application to a new language ecosystem and are then at the mercy of the compiler developers for continued development and support. Developing and maintaining production-quality compilers is challenging and expensive and few organizations can afford such a long-term project.
Processed_DASH:_A_C++_PGAS_Library_for_Distributed_Data_Stru.txt	described in Sect III-B are an example). Global Arrays [11] and OpenSHMEM [12] are two popular examples for compiled PGAS libraries with a C API, which offer an easy integration into existing code bases. However, precompiled libraries and static APIs severely limit the productivity and expressiveness of programming systems and optimizations are typically re- stricted to local inlining of routines.
Processed_DASH:_A_C++_PGAS_Library_for_Distributed_Data_Stru.txt	C++, on the other hand, has powerful abstraction mecha- nisms that allow for generic, expressive, and highly optimized libraries [13]. With a set of long awaited improvements incorporated in C++11 [14], the language has recently been used to implement several new parallel programming systems in projects such as UPC++ [15], Kokkos [16], and RAJA [17]. In this paper we describe DASH, our own C++ template library that implements the PGAS model and provides generic distributed data structures and parallel algorithms. DASH realizes the PGAS model purely as a C++ template library and does not require a custom (pre-)compiler infrastructure, an approach sometimes called compiler-free PGAS. Among the distinctive features of DASH are its inter-operability with existing (MPI) applications, which allows the porting of individual data structures to the PGAS model, and support for hierarchical locality beyond the usual two-level distinction between local and remote data.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Abstract. We discuss and illustrate the behaviour of the continued fraction expansion of a formal power series under specialisation of parameters or their reduction modulo p and sketch some applications of the reduction theorem here proved.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Given a continued fraction expansion over a function ﬁeld K(X) it may happen that a specialisation of incidental parameters leads some partial quotient to acquire inﬁnite coeﬃcients. I point out that in such a case the partial quotient will ‘collapse’ to a partial quotient of higher degree. Indeed, I illustrate several techniques for manipulating continued fraction expansions to display such a ‘collapse’ explicitly. To make sense of the notion “specialisation” the base ﬁeld K = K(t1, t2, . . .) should be a transcendental extension of some yet more base ﬁeld K by one or several algebraically independent parameters t1 , t2 , . . . . In the sequel K will be supposed to be Q or some ﬁnite ﬁeld Fp , but our remarks are often more general. For example, if Y 2 = (X 2 + u)2 + 4v(X + w) viewed as deﬁned over the ﬁeld K , then studying the continued fraction of Y will have us working over K(X), where K = K(u, v, w).
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Deﬁnition 1. A specialisation is a restriction on the generality of the parameters deﬁning K over K .
Processed_Specialisation_and_reduction_of_continued_fraction.txt	For example, taking u + w2 = v above is a specialisation; so is a parametrisation 2t, w = t. A fortiori, a numerical example is a specialisation. u = 1 Our Y above is a formal Laurent series, an element of K((X −1)). Its ‘Taylor’ coeﬃcients are elements of K = K(u, v, w). For simplicity, let me suppose K = Q . I say that replacing the base ﬁeld Q by some ﬁnite ﬁeld, say Fp is a reduction of Y . Of course such a reduction may not make sense, because p occurs in the denominator of some Taylor coeﬃcient. If so, I say variously that Y does not have reduction modulo p or has bad reduction at p.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Furthermore, one sees readily that for the example Y 2 = (X 2 + u)2 + 4v(X + w) reduction modulo 2 either yields Y = X 2 + u , a polynomial or, for specialisations of u , v , w , may make no sense. In either case I may call the reduction bad, in the former case because under reduction the curve decreases in genus.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	rh(X) = xh(X)/yh(X) = [ a0(X) , a1(X) , . . . , ah(X) ]. Now denote by F = P f iX −i some specialisation of F eﬀected by specialising parameters of the ﬁeld over which F is deﬁned, or a reduction of F . Then the sequence of convergents to F is precisely the sequence (cid:0)rh(X)(cid:1), however listed without repetition.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Note here that if some partial quotient ah has no specialisation (in more blunt words, blows up) then neither do the continuants xh and yh . Their quotients rh will nevertheless have a deﬁned specialisation provided only that the specialisation, or reduction, F makes sense. In practice, several distinct convergents rh , rh+1 , . . . , may collapse to the same convergent rh of F .
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Y h=0 are linear if G3 is viewed as deﬁned over F3 they must also be linear for G3 deﬁned over Q . Speciﬁcally, in that example the partial quotients of G3 over Q all have good reduction modulo 3 (as was shown with inappropriate eﬀort in [1]).
Processed_Specialisation_and_reduction_of_continued_fraction.txt	so if rh = xh/yh is a convergent to F then, by (2), chxh/chyh = rh is a convergent to F .
Processed_Specialisation_and_reduction_of_continued_fraction.txt	and necessarily yhyk , has good reduction, proving that there is no convergent to F between the convergents rh and rk , as required.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	The following remarks are intended to show oﬀ several lemmata allowing one to manipulate continued fraction expansions in somewhat surprising ways.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	In particular we will see by manipulation of continued fractions that the blowing up of a partial quotient in a continued fraction expansion leads to a collapse to higher degree. The point is that, even if there is a blowup of partial quotients under specialisation or reduction, one does not have to re-expand the function. In principle one can obtain the new continued fraction expansion directly from the original expansion.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Of course, actually to suggest aX −1 is a partial quotient is “wash your mouth out” stuﬀ and we will have to work hard to make up for the outrage. We begin with several fairly well known lemmata, and their little known corollaries.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	B[ Ca0 , Ba1 , Ca2 , Ba3 , Ca4 , . . . ] = C[ Ba0 , Ca1 , Ba2 , Ca3 , Ba4 , . . . ].
Processed_Specialisation_and_reduction_of_continued_fraction.txt	This fact is both obvious and fairly well known. Its present felicitous formulation is given by Wolfgang Schmidt [14].
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Note that the second expansion (4) contains more information than the ﬁrst (3), whence my reluctance to work from the ﬁrst expansion. Note also that the computation just presented is not the sort of thing one will essay more than once in a lifetime. Indeed, other than over the ﬁnite ﬁeld F2 , and perhaps F3 , it surely cannot be reasonable to attempt to obtain detailed information in this way in general circumstances. Over those ﬁnite ﬁelds, on the other hand, the methods just illustrated may be pursued without pain or fear; for example, there is work of Niederreiter and Vielhaber [8, 9] applying just these notions. Note also the pursuit of the present ideas over Z in a very special case [7], and mention of similar such cases in [10].
Processed_Specialisation_and_reduction_of_continued_fraction.txt	On the other hand, if we were to view D as deﬁned over some ﬁnite ﬁeld, say Fp (with p = 2 ), then necessarily its expansion must be periodic because now there are only ﬁnitely many polynomials of bounded degree. Indeed, inspection of the expansion above shows that the partial quotient a2 blows up at p = 3 and at p = 7 , a5 blows up at p = 5 , a6 at p = 11 , . . . signalling period length r = 2 at p = 3 and p = 7 (more to the point: regulator, the sum of the degrees of the partial quotients making up a quasi-period, m = 3 at those primes), regulator m = 6 at p = 5 , regulator m = 7 at p = 11 , . . . . This more than suﬃces, by a remark of Jing Yu [15], to prove that the expansion of pD(X) over Q is indeed not periodic. Speciﬁcally, Yu points out that, by the reduction theory of abelian varieties, if a divisor class (here, that of the divisor at inﬁnity) is of order m on the Jacobian Jac p , of some curve C ; the Jacobian of the curve C if p divides m then m = mppi , some positive integer i . For our example m7 = 3 and m5 = 6 suﬃces to prove non-periodicity. We should here note that the primes dividing the discriminant of the polynomial X 4 2X 3 + 3X 2 + 2X + 2 are 2 , 3 , and 31 so both 7 and 5 are primes of good reduction.
Processed_Specialisation_and_reduction_of_continued_fraction.txt	See [12] for details or, for diﬀerent methods to construct the families of curves, [6], [2], or [13]; the original tabulation occurs in [5], see also [4].
Processed_Specialisation_and_reduction_of_continued_fraction.txt	Generically, after the ﬁrst several quotients, all partial quotients will be of degree one so, just as in the g = 1 case, periodicity requires specialisations of — that is, relations on — the coeﬃcients f , g , u , v , w . However, now a ‘blowup’ of a partial quotient may mean no more than that a multiple of the divisor at inﬁnity on the Jacobian of the hyperelliptic curve Y 2 = D(X) is (technically, corresponds to) a point on the curve and, unlike the elliptic case, does not guarantee periodicity. It will be interesting to see whether the Reduction Theorem is helpful in practice in recognising blowup to degree g + 1 .
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	ABSTRACT We present optical, radio and X-ray data that conﬁrm a new supernova remnant (SNR) in the Large Magellanic Cloud (LMC) discovered using our deep Hα imagery. Optically, the new SNR has a somewhat ﬁlamentary morphology and a diameter of 56 × 64 arcsec (13.5 × 15.5 pc at the 49.9 kpc distance of the LMC). Spectroscopic follow-up of multiple regions show high [S ii]/Hα emission-line ratios ranging from 0.66 ±0.02 to 0.93 ±0.01, all of which are typical of an SNR. We found radio counterparts for this object using our new Australia Telescope Compact Array (ATCA) 6 cm pointed observations as well as a number of available radio surveys at 8 640 MHz, 4 850 MHz, 1 377 MHz and 843 MHz. With these combined data we provide a spectral index α ≈ – 0.5 between 843 and 8 640 MHz. Both spectral line analysis and the magnetic ﬁeld strength, ranging from 124 − 184 µG, suggest a dynamical age between ∼2,200 and ∼4,700 yrs. The SNR has a previously catalogued X-ray counterpart listed as HP 483 in the ROSAT Position Sensitive Proportional Counter (PSPC) catalogue.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	the ultraviolet (Blair et al. 2006). Multi-wavelength studies of SNRs provide information on several crucial aspects of stellar evolution and the physical behaviour of shocked ma- terial. They have a major inﬂuence on local magnetic ﬁelds whilst the shock waves may trigger new star formation and accelerate cosmic rays.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	Figure 1. The median stacked and combined Hα and broad band red optical image of RP1577 where Hα is shown as red and the short red is shown as white.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	of [S ii] to Hα > 0.5 implies a shock front indicative of an SNR as does strong [O i] and [O ii] lines plus the supporting presence of [N ii] (Sabbadin et al. 1977). In addition, the de- tection of Hα, rather than [O iii], leading the radiation front would support the rejection of any Wolf-Rayet (WR) shell scenario (Gruendl et al. 2000).
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	At a distance of ∼49.9 kpc (Reid & Parker 2010 and references therein) the LMC is an ideal laboratory to study the diﬀerent varieties of SNRs in great detail across an entire galaxy. The LMC is close enough that SNRs are resolved suf- ﬁciently for multiwavelength studies to be conducted. Since SNRs yield key information on the death rate of high mass stars and the number of white dwarfs exploding as Type Ia SN, an improved inventory of LMC SNRs, complete as possible, is required to put constraints on these values and improve our understanding of the ﬁnal stages of stellar evo- lution. The LMC SNR sample is currently estimated at 59 remnants1.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	The deep, stacked UKST Hα survey used to search for planetary nebulae in the LMC (Reid & Parker, 2006a, 2006b, 2013), also revealed several candidate SNRs, the ﬁrst of which we report here, thanks to the optical depth of the Hα survey to Requiv mag 22, and the method of image anal- ysis. Spectroscopic followup of these objects (originally iden- tiﬁed as SNR candidates in 2004) using 2dF on the Anglo- Australian Telescope (AAT) and the 1.9-m telescope at the South African Astronomical Observatory (SAAO) have per- mitted us to conﬁrm identiﬁcations and conduct an optical spectroscopic and multiwavelength analysis.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	The rest of the paper is structured as follows. In sec- tion 2 we present the optical observational data including images and spectra. In subsection 2.1 we derive an estimate for the expansion velocity and age of the SNR and in 2.2 estimate the radial velocity from key emission lines. In sec- tion 3 we provide supporting radio data, the results of in- frared imaging and previous X-ray observations. Section 4 is our conclusion where we summarize the combined multi- wavelength data used to achieve our classiﬁcation.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	The new SNR candidate, given the Reid & Parker refer- ence name RP1577 and presented here for the ﬁrst time, was discovered using deep-stack Hα and short red images of the LMC, SMC and surrounding regions (Reid & Parker 2006a). An Hα interference ﬁlter of exceptional speciﬁcation and quality (Parker & Bland-Hawthorn 1998) was used to cover a substantial fraction of the UKST’s large ﬁeld. Its central wavelength (6590 ˚A) and bandpass (70 ˚A FWHM) work eﬀectively in the UKST’s fast f/2.48 converging beam. With a peak transmission of >85 per cent, the Hα ﬁlter is also partially sensitive to [N ii]6583 ˚A. Overlapping ﬁelds with non-standard 4-degree centres enabled full contiguous coverage in Hα despite the circular aperture. High resolu- tion, panchromatic Tech-Pan ﬁlm with peak sensitivity at Hα was used as the photographic detector (see Parker et al. 2005 for more details).
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	Over a period of three years, from 1997, a series of 12 repeated narrow-band Hα and 6 matching broad-band ‘SR’ (Short Red) exposures of the central LMC ﬁeld were also taken (see Reid & Parker 2006a). The purpose was that these images should be stacked to produce a single Hα and short red image with an overall depth increase of 1 magnitude over a single image frame. The twelve highest quality and well- matched UK Schmidt Telescope 2-hour Hα exposures and six 15-minute equivalent SR-band exposures were selected. From these exposures, deep, homogeneous, narrow-band Hα and matching broad-band SR maps of the entire central 25 deg2 square of the LMC were constructed.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	The ‘SuperCOSMOS’ plate-measuring machine at the Royal Observatory Edinburgh (Hambly et al. 2001) scanned, co-added and pixel matched these exposures creating 10µm (0.67 arcsec) pixel data which goes 1.35 and 1 magnitudes deeper than individual exposures, achieving the full canon- ical Poissonian depth gain, e.g. Bland-Hawthorn, Shopbell & Malin (1993). This gives a depth ∼21.5 for the SR images and Requiv ∼22 for Hα (4.5 × 10−17ergs cm−2 s−1 ˚A−1) which is at least 1-magnitude deeper than the best wide-ﬁeld narrow-band LMC images previously available. An accurate world co-ordinate system was applied to yield sub-arcsec as- trometry, essential for success of the spectroscopic follow-up observations.
Processed_Optical_discovery_and_multiwavelength_investigatio.txt	Table 1. Observing Logs for the follow-up of emission object and SNR candidate RP1577 in the LMC. The ﬁrst column gives the telescope name, second gives the instrument name, third gives the observation date, the fourth gives the grating name, the ﬁfth gives the dispersion in angstroms, the sixth gives the central wavelength, the seventh gives the wavelength coverage in angstroms, the eighth gives the exposure time in seconds, the ninth gives the number of exposures taken.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.