A Comparative Simulation Study Of The Heteroscedasticity Consistent Covariance Matrix Estimators In The Linear Regression Model

Statistics Project Topics

Get the Complete Project Materials Now! ยป

III the context of econometric methods of estimation the variances of OLS estimatesrnderived under the assumption of homoscedasticity are not consistent when there isrnheteroscedasticity and their use can lead to incorrect inferences. Thus, this paper sets outrnto examine the performance of several modified versions of heteroscedasticity consistentrncovariance matrix (HCCM) estimator (namely HCO, HC I , HC2, and HC3) of Whitern(1980) and white and MackiJU10n (1985) over a range of sample sizes. Most applicationsrnthat use HCCM appear to rely on HCO, yet tests based on the other HCCM estimators arernfound to be consistent even in the presence of heteroscedasticity of an unknown form .rnBased on Monte Carlo experiments which compare the performance of the t statistic, itrnwas found out that HC2 and HC3 estimators precisely out perform the others in smallrnsamples. In particular HC3 estimator for samples of size less than 100 was found to bernbetter than the other HCCM estimators; when samples are 250 or larger, other versions ofrnthe HCCM can be used. Added to that, it was cost advantageous to employ HC3 insteadrnof ordinary least square covariance matrix (OLSCM) even when there is li ttle evidence ofrnhetreoscedastici ty.rnKey wordsrnWhite estimator, Monte Carlo Simulation, Linear Regression, Heterosexuality

Get Full Work

Report copyright infringement or plagiarism

Be the First to Share On Social



1GB data
1GB data

RELATED TOPICS

1GB data
1GB data
A Comparative Simulation Study Of The Heteroscedasticity Consistent Covariance Matrix Estimators In The Linear Regression Model

134