Activity Number:
|
176
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Business & Economics Statistics Section*
|
Abstract - #300720 |
Title:
|
A Test for the Equality of Regression Coefficients when the Regressions are Heteroscedastic
|
Author(s):
|
Dennis Oberhelman*+ and Rao Kadiyala
|
Affiliation(s):
|
University of South Carolina and Purdue University
|
Address:
|
1705 College Street, Columbia, South Carolina, 29208, USA
|
Keywords:
|
Behrens Fisher Problem ; Heteroscedasticity ; Linear Model ; Hypothesis Test
|
Abstract:
|
Testing for the equality of regression coefficients across two regressions is a problem considered by analysts in a variety of fields. If the variances of the errors of the two regressions are not equal, then it is known that the standard F-test used to test the equality of the coefficients performs poorly in that its actual size can differ substantially from the stated level of significance. This paper addresses this problem and borrows from the literature on the Behrens-Fisher problem to develop an easily applied test whose actual size is close to the stated level of significance. Monte Carlo results indicate that the proposed test is superior to well-known alternative tests over a wide range of the parameter space.
|