So, I took the sample test yesterday and one of the questions got me confused and I don’t remember seeing this concept anywhere in the books. Basically, the question asked whether there was Unit Root and/or Heteroskedasticity in the regression model. The Unit Root T-Test was very high (about 16 or 17) and the critical value for Unit Root was 1.96. So, we should reject the hypothesis that there is a unit root. However, for Heteroskedasticity, the T-Test was about 2.11, while the critical value was 1.96. For some reason, the correct was that there IS heteroskedasticity in the regression. Could someone please explain this??? To me, it seems like since 2.11 is greater than the critical value of 1.96, we should reject the hypothesis that there is heteroskedasticity in the regression. Maybe I’m just missing something… Also, where can one find the unit root test and the heteroskedasticity test in the program?
I know that question tripped me up too and the explanation wasn’t very helpful either. Esp if they ONLY give you the t-tests and the critical t-test. They had heteroskedascitity info earlier in the item set, but they didn’t say to refer to it, they only said to refer to the other problem. Usually, when they point you to one exhibit, you’d assume you only need that one exhibit.
Thanks for the info. I did not even see it there. Do you by any chance know if the heteroskedasticity test is performed the same way? If the Chi-Square is larger than the critical value at a certain significance level, then we reject the hypothesis that there is heteroskedasticity?
I don’t think this concept was talked about much in the program. Does anyone have the page number for the CFA books?