Regression estimation

From Wikipedia, the free encyclopedia

Regression estimation is a technique used to replace missing values in data. The variable with missing data is treated as the dependent variable, while the rest of the cases are treated as independent variables. A regression equasion is then generated which can be used to predict missing values. This method reduces the variance associated with other techniques.

References Tabachnick, B. G., & Fidel, L. S. (2001). Using multivariate statistics (4th ed.). Boston, Mass. : Allyn and Bacon.