Accession Number : AD0635884

Title :   LOG-LINEAR REGRESSION.

Corporate Author : GEORGE WASHINGTON UNIV WASHINGTON D C LOGISTICS RESEARCH PROJECT

Personal Author(s) : Heien, Dale M.

Report Date : 18 JUL 1966

Pagination or Media Count : 13

Abstract : The paper presents a reparameterization of the conventional log-linear regression model. The Stochastic assumptions of this reparameterized model are derived from the expected value properties of the dependent variable. Comparison with the conventional method of fitting log-linear models shows the regression intercept and the variance of the regression intercept to be different. A t-test is proposed for the intercept. Also regression through the origin is treated briefly. (Author)

Descriptors :   (*STATISTICAL ANALYSIS, STOCHASTIC PROCESSES), STATISTICAL DISTRIBUTIONS

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE