On modified unbiased two-parameter estimator

Abstract

The ridge-type estimators have been intensively studied and modified for the linear regression model. In this article, we introduce a modified unbiased two-parameter estimator (MUTPE) as a new estimator to solve the multicollinearity problem for the linear regression model. The MUTPE has been obtained as a linear combination of unbiased two-parameter estimator (UTPE). We give a simulation study to demonstrate the theoretical results. The results of the simulation have revealed that the proposed estimator has better effectiveness than both UTPE and ridge estimators under some circumstance. Finally, we analyzed a real-life data to justify the performance of the modified estimator MUTPE in the context of a linear regression model.

Keywords:

• Linear regression model
• multicollinearity
• two-parameter estimator
• unbiased