Analyzing the Distribution of Pre-test Estimator of Error Variance in Linear Regression Model

Abstract

In statistical inference, the estimation of error variance plays a crucial role in regression analysis and hypothesis testing. Using the available extraneous information, some improved estimators of error variance in the linear regression model may be constructed which combine both the sample and the non-sample information. A pre-test estimator of error variance is considered when prior information about regression parameters is available, influencing the estimation process. The modest aim of this article is to study and analyze distributional properties of a pre-test estimator of error variance in a linear regression model with numerical computation.