Title: Computer-intensive methods for dependent
data
Description:
The development of theoretical and applied
statistics has been widely influenced by the availability of powerful and cheap
computers. Numerical calculations have become progressively less bundersome with
the increasing speed of computers, while at the same time, it has been possible
to use larger data sets. Moreover, since the seminal paper by Efron on the
bootstrap, it was clear that completely new statistical techniques, that involve
powerful computer calculations, could greatly extend the statistical analysis of
complex problems with complicated statistical models. It was soon realized that
it was possible to provide statistical tools that works in complex situations
without imposing unrealistic and unverifiable assumptions about the data
generating mechanism. In other words, it was possible to weaken underlying
assumptions by means of raw computing power. After a decade of proposals and
tinkering, statisticians have agreed on the theoretical importance and the
practical use of resampling, nonparametric and simulation- based techniques to
determine more accurately the reliability of data analysis in a wide choice of
subjects ranging from politics to medicine to particle physics.
From the computationally point of view, a new computing breakthrough is
approaching with accessibility of parallel and grid computing and availability
of packages to implement parallel algorithms in popular languages (such as Ox or
R). This will give a new burst to simulation-based techniques, such as the
bootstrap and the subsampling, that do not need re-definition of the computing
algorithms, being intrinsically parallel techniques.
We will cover new statistical and computational issues on the proposed topic
with an emphasis on dependent data.
Focus:
Bootstrap
Subsampling
Non-parametric techniques
Empirical likelihood
Simulation-based methods (Indirect-inference and related methods)
Parallel computing
Co-Chairs:
Rong Chen
Peking University
Dept. of Info. and Decision Sciences (M/C 294)
College of Business Administration
The University of Illinois at Chicago
601 Morgan Street
Chicago, IL 60607-7124
Tel: (312) 996 2323
E-mail: rongchen@uic.edu
|
Michele La Rocca
Dept. of Economics and Statistics
University of Salerno
Via Ponte don Melillo
84084 Fisciano (SA)
Italy
tel. +39 089-962200
fax +39 089-962049
E-mail: larocca@unisa.it |
Dimitris N. Politis
Department of Mathematics
University of California, San Diego
La Jolla, CA 92093-0112
USA
Tel.: +1 858 534-5861
Fax: +1 858 534-5273
E-mail: dpolitis@ucsd.edu
|
Qiwei Yao
London School of Economics
Houghton Street
London, WC2A 2AE
United Kingdom
Tel: +44 (0)20 7955 6767
Fax: +44 (0)20 7955 7416
E-mail: q.yao@lse.ac.uk
|
|