Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1149780 | Journal of Statistical Planning and Inference | 2009 | 14 Pages |
Abstract
A standard assumption in regression analysis is homogeneity of the error variance. Violation of this assumption can have adverse consequences for the efficiency of estimators. In this paper, we propose an empirical likelihood based diagnostic technique for heteroscedasticity in the partially linear errors-in-variables models. Under mild conditions, a nonparametric version of Wilk's theorem is derived. Simulation results reveal that our test performs well in both size and power.
Keywords
Related Topics
Physical Sciences and Engineering
Mathematics
Applied Mathematics
Authors
Heung Wong, Feng Liu, Min Chen, Wai Cheung Ip,