Publication:
On the asymptotic theory of subsampling

Thumbnail Image
Identifiers
URI: http://hdl.handle.net/10016/6334
Files
ws995217.PDF (1.53 MB)
Publication date
1999-06
Defense date
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Google Scholar
Export
Research Projects
Organizational Units
Journal Issue
Abstract
A general approach to constructing confidence intervals by subsampling was presented in Politis and Romano (1994). The crux of the method is based on recomputing a statistic over subsamples of the data, and these recomputed values are used to build up an estimated sampling distribution. The method works under extremely weak conditions, it applies to independent, identically distributed (LLd.) observations as well as to dependent data situations, such as time series (possible non stationary) , random fields, and marked point processes. In this article, we present some new theorems showing: a new construction for confidence intervals that removes a previous condition, a general theorem showing the validity of subsampling for datadependent choices of the block size, and a general theorem for the construction of hypothesis tests (which is not necessarily derived from a confidence interval construction). The arguments apply to both the Li.d. setting as well as the dependent data case.
Description
Keywords
Confidence intervals, Data--dependent block size choice, Hypothesis tests, large sample theory, resampling
Bibliographic citation
Collections
DES - Working Papers. Statistics and Econometrics. WS