ON THE ISSUE OF INTERPRETING COEFFICIENTS OF LINEAR REGRESSION

UDC 330.43 (075.8)
DOI: 10.36871/2618-9976.2024.01.001

Authors

Irina V. Orlova,
Candidate of Economic Sciences, Professor, Financial University under the Government of the Russian Federation, Moscow, Russia
Daria P. Orlova,
Financial University under the Government of the Russian Federation, Moscow, Russia

Abstract

The paper considers two approaches to estimating the change in the endogenous variable Y when changing some regressor Xj by one. It is shown how different values of this estimate for different approaches are associated with differences in views on accounting for the correlation dependencies of regressors, coefficients are proposed to calculate the relative contribution of each regressor to the values of estimates of regression coefficients and covariance between regressors and Y. The results of the study are illustrated by a model example.

Keywords

Linear regression, Covariance, Interpretation of regression coefficients