Weighted modified first order regression procedures for estimation in linear models with missingX-observations

Abstract

This paper considers the estimation of coe cients in a linear regression model with missing observations in the independent v ariables and introduces a modi cation of the standard rst order regression method for imputation of missing values. The modi cation provides stochastic values for imputation and, as an extension, makes use of the principle of weighted mixed regression. The proposed procedures are compared with two popular procedures|one which utilizes only the complete observations and the other which employs the standard rst order regression imputation method for missing values.A simulation experiment t o e v aluate the gain in e ciency and to examine interesting issues like the impact of varying degree of multicollinearity in explanatory variables is proceeded. Some work on the case of discrete regressor variables is in progress and will be reported in a future article to follow.