The Effect of Different Imputation Methods on Analytical Statistics of Simple Linear Regression
by Jann-Huei Jinn
We have considered the simple linear regression model
where the errors are normally and independently distributed with E( )=0 and Var( )= . Let , and denote the estimates of , , and where "c" indicates that these are estimates from a completed data set. We present the results of a theoretical investigation of the effect that commonly used imputation methods have on the properties of confidence for and . Specifically, we will try to identify "good" imputation procedures and conditions where specific imputation procedures work well.
Simple Linear Regression, Imputation
READING THE ARTICLE: You can read the article in
portable document (.pdf) format (244625 bytes.)
NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.
This page has been accessed 3092 times since July 24, 2006.
Return to the Home Page.