Application of Stein-Rule Estimation to Linear Regression Models with Some Missing Observations

The problem of estimating the coefficients in a linear regression model is considered when some of the response values are missing. The conventional Yates procedure employing least squares predictions for missing values does not lead to any improvement over the least squares estimator using complete observations only. However, if we use Stein-rule predictions, it is demonstrated that some improvement can be achieved. An unbiased estimator of the mean squared error matrix of the proposed estimator of coefficient vector is also presented. Some work on the application of the proposed estimation procedure to real-world data sets involving some discrete variables in the set of explanatory variables is under way and will be reported in future. © 2006, The Indian Econometric Society.