Abstract
We study the finite-sample properties of White's test for heteroskedasticity in stochastic regression models where explanatory variables are random and not given. We investigate by simulation the effect of non independence of explanatory variables and error term and heteroskedasticity on White's test. A standard bootstrap method in the computationally convenient form is found to work well with respect to the size and power.
Mathematics Subject Classification:
Acknowledgment
The authors thank helpful comments by an anonymous referee. Masakazu Ando is financially supported for this work by Grant-in-Aid for Scientific Research (KAKENHI(16.12163)).
Notes
1We remark the heteroskedasticity consistent covariance matrix estimators such as HC1, HC2, HC3, and HC4 (cf., e.g., Hodoshima and Ando, Citation2006) do not work for White's test for heteroskedasticity since they do not satisfy the condition (cf., the proof of Theorem 2 of White, Citation1980) when
is replaced by the heteroskedasticity consistent covariance matrix estimators.