# Breusch-Pagan test for heteroscedasticity contradicts White's test?

Testing for heteroscedasticity I get these results:

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

• $H_0$: Constant variance
• $H_a$: Heteroskedasticity

 Variables: fitted values of log_expdu

chi2(1)      =    21.41
Prob > chi2  =   0.0000


White's test

• $H_0$: homoskedasticity

• against $H_a$: unrestricted heteroskedasticity

 chi2(47)     =     48.91
Prob > chi2  =    0.3964


The Breusch-Pagan test tells me that there is heteroskedasticity, while the White's test tells me the opposite. Which result should I use?

The Breusch-Pagan test only checks for the linear form of heteroskedasticity i.e. it models the error variance as $\sigma_i^2 = \sigma^2h(z_i'\alpha)$ where $z_i$ is a vector of your independent variables. It tests $H_0: \alpha = 0$ versus $H_a: \alpha \neq 0$.