Robustness of size of tests of autocorrelation and heteroscedasticity to nonnormality

Merran Evans

doi:10.1016/0304-4076(92)90027-O

Robustness of size of tests of autocorrelation and heteroscedasticity to nonnormality

Merran Evans

Econometrics & Business Statistics

Research output: Contribution to journal › Article › Research › peer-review

24 Citations (Scopus)

Abstract

A comprehensive empirical examination is made of the sensitivity of tests of disturbance covariance in the linear regression model to nonnormal disturbance behaviour. Tests of autocorrelation appear to be quite robust, except for extreme nonnormality, but tests for heteroscedasticity are highly susceptible to kurtosis.

Original language	English
Pages (from-to)	7-24
Number of pages	18
Journal	Journal of Econometrics
Volume	51
Issue number	1-2
DOIs	https://doi.org/10.1016/0304-4076(92)90027-O
Publication status	Published - 1992

Access to Document

10.1016/0304-4076(92)90027-O

Cite this

@article{40f25a420bab45678f14ce170b21d05c,

title = "Robustness of size of tests of autocorrelation and heteroscedasticity to nonnormality",

abstract = "A comprehensive empirical examination is made of the sensitivity of tests of disturbance covariance in the linear regression model to nonnormal disturbance behaviour. Tests of autocorrelation appear to be quite robust, except for extreme nonnormality, but tests for heteroscedasticity are highly susceptible to kurtosis.",

author = "Merran Evans",

year = "1992",

doi = "10.1016/0304-4076(92)90027-O",

language = "English",

volume = "51",

pages = "7--24",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier",

number = "1-2",

}

TY - JOUR

T1 - Robustness of size of tests of autocorrelation and heteroscedasticity to nonnormality

AU - Evans, Merran

PY - 1992

Y1 - 1992

N2 - A comprehensive empirical examination is made of the sensitivity of tests of disturbance covariance in the linear regression model to nonnormal disturbance behaviour. Tests of autocorrelation appear to be quite robust, except for extreme nonnormality, but tests for heteroscedasticity are highly susceptible to kurtosis.

AB - A comprehensive empirical examination is made of the sensitivity of tests of disturbance covariance in the linear regression model to nonnormal disturbance behaviour. Tests of autocorrelation appear to be quite robust, except for extreme nonnormality, but tests for heteroscedasticity are highly susceptible to kurtosis.

UR - https://www.scopus.com/pages/publications/38249015963

U2 - 10.1016/0304-4076(92)90027-O

DO - 10.1016/0304-4076(92)90027-O

M3 - Article

AN - SCOPUS:38249015963

SN - 0304-4076

VL - 51

SP - 7

EP - 24

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 1-2

ER -

Robustness of size of tests of autocorrelation and heteroscedasticity to nonnormality

Abstract

Access to Document

Other files and links

Cite this