Scott Anthony Sisson

Scott Anthony Sisson

Auto-generated profile, based on publicly available information

Scott Anthony Sisson

School of Mathematics and Statistics, University of New South Wales

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 February 2007 in Proceedings of the National Academy of Sciences
Verified
Journal Article
Closed access
SSS. A. Sisson
S. A. Sisson
YFY. Fan
Y. Fan
MTMark M. Tanaka
Mark M. Tanaka
Recent new methods in Bayesian simulation have provided ways of evaluating posterior distributions in the presence of analytically or computationally intractable likelihood functions. Despite representing a substantial methodological advance, existing methods based on rejection sampling or Markov chain Monte Carlo can be highly inefficient and accordingly require far more iterations than may be practical to implement. Here we propose a sequential Monte Carlo sampler that convincingly overcomes these inefficiencies. We demonstrate its implementation through an epidemiological study of the transmission rate of tuberculosis.

1 March 2010 in Journal of the American Statistical Association
Verified
Journal Article
Closed access
SPS. A. Padoan
S. A. Padoan
MRM. Ribatet
M. Ribatet
SSS. A. Sisson
S. A. Sisson
The last decade has seen max-stable processes emerge as a common tool for the statistical modeling of spatial extremes. However, their application is complicated due to the unavailability of the mu...

1 May 2013 in Statistical Science
Verified
Journal Article
Open access
PDF file
MBM. G. B. Blum
M. G. B. Blum
MNM. A. Nunes
M. A. Nunes
DPD. Prangle
D. Prangle
SSS. A. Sisson
S. A. Sisson
Approximate Bayesian computation (ABC) methods make use of comparisons between simulated and observed summary statistics to overcome the problem of computationally intractable likelihood functions. As the practical implementation of ABC requires computations based on vectors of summary statistics, rather than full data sets, a central question is how to derive low-dimensional summary statistics from the observed data with minimal loss of information. In this article we provide a comprehensive review and comparison of the performance of the principal methods of dimension reduction proposed in the ABC literature. The methods are split into three nonmutually exclusive classes consisting of best subset selection methods, projection techniques and regularization. In addition, we introduce two new methods of dimension reduction. The first is a best subset selection method based on Akaike and Bayesian information criteria, and the second uses ridge regression as a regularization procedure. We illustrate the performance of these dimension reduction techniques through the analysis of three challenging models and data sets.

1 March 2003 in Journal of Hydrology
Verified
Journal Article
Closed access
SCStuart Coles
Stuart Coles
LPLuis Pericchi
Luis Pericchi
SSScott Sisson
Scott Sisson

1 November 2013 in Journal of Hydrology
Verified
Journal Article
Closed access
FZFeifei Zheng
Feifei Zheng
SWSeth Westra
Seth Westra
SSScott A. Sisson
Scott A. Sisson

24 February 2014 in Water Resources Research
Verified
Journal Article
Open access
PDF file
FZFeifei Zheng
Feifei Zheng
SWSeth Westra
Seth Westra
MLMichael Leonard
Michael Leonard
SSScott A. Sisson
Scott A. Sisson
Accounting for dependence between extreme rainfall and storm surge can be critical for correctly estimating coastal flood risk. Several statistical methods are available for modeling such extremal dependence, but the comparative performance of these methods for quantifying the exceedance probability of rare coastal floods is unknown. This paper compares three classes of statistical methods—threshold‐excess, point process, and conditional—in terms of their ability to quantify flood risk. The threshold‐excess method offers approximately unbiased estimates for dependence parameters, but its application for quantifying flood risk is limited because it is unable to handle situations where only one of the two variables is extreme. In contrast, the point process method (with the logistic and negative logistic models) and the conditional method describe the full distribution of extremes, but they overestimate and underestimate the dependence strength, respectively. We conclude that the point process method is the most suitable approach for modeling dependence between extreme rainfall and storm surge when the dependence is relatively strong, while none of the three methods produces satisfactory results for bivariate extremes with very weak dependence. It is therefore important to take the bias of each method into account when applying them to flood estimation problems. A case study is used to demonstrate the three statistical methods and illustrate the implication of dependence to flood risk.