When are first order asymptotics adequate? A diagnostic

K A Anaya Izquierdo, Frank Critchley, Paul Marriott

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This paper looks at boundary effects on inference in an important class of models including, notably, logistic regression. Asymptotic results are not uniform across such models. Accordingly, whatever their order, methods asymptotic in sample size will ultimately ’break down’ as the boundary is approached, in the sense that effects
such as infinite skewness, discreteness and collinearity will dominate. In this paper, a highly interpretable diagnostic tool is proposed allowing the analyst to check if the boundary is going to have an appreciable effect on standard inferential techniques.
Original languageEnglish
Pages (from-to)17-22
Number of pages6
JournalStat
Volume3
Issue number1
Early online date21 Feb 2014
DOIs
Publication statusPublished - Feb 2014

Fingerprint

Diagnostics
First-order
Collinearity
Boundary Effect
Asymptotic Methods
Skewness
Logistic Regression
Breakdown
Sample Size
Model
Class
Standards

Cite this

When are first order asymptotics adequate? A diagnostic. / Anaya Izquierdo, K A; Critchley, Frank; Marriott, Paul.

In: Stat, Vol. 3, No. 1, 02.2014, p. 17-22.

Research output: Contribution to journalArticle

Anaya Izquierdo, K A ; Critchley, Frank ; Marriott, Paul. / When are first order asymptotics adequate? A diagnostic. In: Stat. 2014 ; Vol. 3, No. 1. pp. 17-22.
@article{c2730f6ebe8241a99d8da5f224094699,
title = "When are first order asymptotics adequate? A diagnostic",
abstract = "This paper looks at boundary effects on inference in an important class of models including, notably, logistic regression. Asymptotic results are not uniform across such models. Accordingly, whatever their order, methods asymptotic in sample size will ultimately ’break down’ as the boundary is approached, in the sense that effectssuch as infinite skewness, discreteness and collinearity will dominate. In this paper, a highly interpretable diagnostic tool is proposed allowing the analyst to check if the boundary is going to have an appreciable effect on standard inferential techniques.",
author = "{Anaya Izquierdo}, {K A} and Frank Critchley and Paul Marriott",
year = "2014",
month = "2",
doi = "10.1002/sta4.40",
language = "English",
volume = "3",
pages = "17--22",
journal = "Stat",
issn = "2049-1573",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - When are first order asymptotics adequate? A diagnostic

AU - Anaya Izquierdo, K A

AU - Critchley, Frank

AU - Marriott, Paul

PY - 2014/2

Y1 - 2014/2

N2 - This paper looks at boundary effects on inference in an important class of models including, notably, logistic regression. Asymptotic results are not uniform across such models. Accordingly, whatever their order, methods asymptotic in sample size will ultimately ’break down’ as the boundary is approached, in the sense that effectssuch as infinite skewness, discreteness and collinearity will dominate. In this paper, a highly interpretable diagnostic tool is proposed allowing the analyst to check if the boundary is going to have an appreciable effect on standard inferential techniques.

AB - This paper looks at boundary effects on inference in an important class of models including, notably, logistic regression. Asymptotic results are not uniform across such models. Accordingly, whatever their order, methods asymptotic in sample size will ultimately ’break down’ as the boundary is approached, in the sense that effectssuch as infinite skewness, discreteness and collinearity will dominate. In this paper, a highly interpretable diagnostic tool is proposed allowing the analyst to check if the boundary is going to have an appreciable effect on standard inferential techniques.

UR - http://onlinelibrary.wiley.com/doi/10.1002/sta4.40/abstract

UR - http://dx.doi.org/10.1002/sta4.40

U2 - 10.1002/sta4.40

DO - 10.1002/sta4.40

M3 - Article

VL - 3

SP - 17

EP - 22

JO - Stat

JF - Stat

SN - 2049-1573

IS - 1

ER -