HUI Working Papers,
HUI Research

Abstract: Despite that interaction terms are standard tools of regression analysis, the side effects of the inclusion of these terms in models estimated by ordinary least squares (OLS) are yet not fully penetrated. The inclusion of interaction effects induces multicollinearity problems since all non-zero values are equal between the interaction term and the regressor. In this article we propose a procedure to remedy this problem by the use of new ridge regression (RR) shrinkage parameters – which we call the asymmetric interaction ridge (AIR) regression method. By means of Monte Carlo simulations we evaluate both OLS and AIR using the mean square error (MSE) performance criterion. The result from the simulation study confirms our hypothesis that AIR always should be preferred to OLS since it has a lower estimated MSE. Moreover, the advantages of our new method is demonstrated in an empirical application where positive asymmetric price transmission effects are exposed for the mortgage interest rates of Handelsbanken Stadshypotek. It is observed that the mortgage interest rates increase more fully and rapidly to an increase in the bank’s borrowing costs than to a decrease. This asymmetry is defined as positive asymmetric price transmission (APT).

Keywords: OLS; multicollinearity; regression analysis

JEL-codes: C30

19 pages, February 15, 2012

Full text files

MediaBinaryLoader.ax...e_ForceDownload=true  

Download statistics

• HUI Research
• Ordering Working Papers
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to Hans Seerar Westerberg ()
Report other problems with accessing this service to Sune Karlsson ().

RePEc:hhs:huiwps:0054This page generated on 2024-09-13 22:15:10.