I am trying to do a Breusch pagan test and I get for two regressions with different dependent variables but with the same independent variables different p-values. Actually, I thought that in the Breusch pagan test, only the independent variables matter and if they are the same in both models, then the p-value should also be the same. Am I right? Why didn't I get for both models the same p-value?
model1.1=lm(giniA~region+dummy_2009,data=fors) model2.1=lm(giniE~region+dummy_2009,data=fors) bptest(model1.1) bptest(model2.1)
model1.1 --> p-value = 0.001086
model2.1 --> p-value = 3.71e-05
Thank you for your help.