[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: svyreg and heteroscedasticity tests
-hettest- does not work when following a -svyreg- command but does work following a -reg y x [w=pweight]- command, which generates exactly the same residuals as the svyreg command.
My interest is in testing the hypothesis that a model estimated on the population would have homoscedastic disturbances but the only data available for doing this test is a complex sample survey.
Is it correct to infer that such a test strategy is invalid, given that -hettest- does not work after a svyreg?
Department of Economics
University of Waikato
* For searches and help try: