Edgeworth Expansions in Statistics: A Brief Review.
Abstract
Asymptotic expansions for distribution functions are of great importance in statistical inference. The basic aspects and some recent developments regarding Edgeworth expansions and Cornish-Fisher expansions are discussed. In recent years, substantial progress has been made in obtaining Berry-Esseen bounds and Edgeworth expansions for test statistics such as linear rank statistics, U-statistics, linear (trimmed and untrimmed) combinations of order statistics. An account of these results is given in Section 4 followed by expansions for minimum contrast estimators and Fisher-consistent estimators.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1982
- Accession Number
- ADA117061
Entities
People
- S. Panchapakesan
- Shanti Gupta
Organizations
- Purdue University