Linear Regression Analysis With Missing Observations Among the Independent Variables
- 1 September 1964
- journal article
- research article
- Published by JSTOR in Journal of the American Statistical Association
- Vol. 59 (307) , 834
- https://doi.org/10.2307/2283101
Abstract
An intuitive estimate is presented for estimating linear regression coefficients when an unspecified number and pattern of the independent variables for some individuals are not observed. It is shown that the efficiency of this approach depends upon the correlation among the independent variables as well as the proportion of observations which is missing.Keywords
This publication has 0 references indexed in Scilit: