UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Gross Non-Normality and the Quality of a Simple Approximation to the P-Value of a Routine Test of Non-Nested Regressions

Szroeter, J; (1998) Gross Non-Normality and the Quality of a Simple Approximation to the P-Value of a Routine Test of Non-Nested Regressions. (UCL Economics Discussion Paper 1998-5). Green open access

[thumbnail of 18123.pdf]
Preview
PDF
18123.pdf

Download (78kB)

Abstract

The distribution of certain test statistics for non-nested regressions can be so grossly non-normal that p-values computed on the assumption of approximate normality cannot be safely used for routine inference. This paper presents results on the quality of a new more accurate yet still user-friendly p-value approximation which embodies an inverse measure of the strength of relationship between regressors of competing models. This easily-computed measure is equivalent to the sum of eigenvalues which have recently been shown to characterize the exact finite-sample distribution of the test statistic.

Type: Working / discussion paper
Title: Gross Non-Normality and the Quality of a Simple Approximation to the P-Value of a Routine Test of Non-Nested Regressions
Open access status: An open access version is available from UCL Discovery
Keywords: non-nested regressions, p-value, size, non-normality, point approximation, eigenvalues.
UCL classification: UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL SLASH
UCL > Provost and Vice Provost Offices > UCL SLASH > Faculty of S&HS > Dept of Economics
URI: https://discovery.ucl.ac.uk/id/eprint/18123
Downloads since deposit
184Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item