[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Test for panel-level heteroskedasticity - what is the hypothesis
At 03:17 PM 8/11/2005 +0200, you wrote:
I'm testing for panel-level heteroskedasticity and I apply the lrtest
suggested in the FAQ section of Stata
homoskedasticity is the null. In the alternative, you are less restrictive
because you allow error variances to vary.
I obtain the LR test result at the end but my problem is that I'm not
really sure about the hypothesis I'm testing. Is the null hypothesis equal
to "panels' error structures are heteroskedastic" or "panels' error
structures are homoskedastic"?
Also, note how the FAQ is worded. The second model is the one "without
heteroskedasticity". Further, it says "The number of panels/groups is
stored in e(N_g) and, in the second model, we are constraining all of these
to be single value". So, the 2nd model is the constrained model and hence
represents the null hyp.
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try: