[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: -test- after -anova , sequential-
Is there a reason why -test- after -anova , sequential- uses partial sums of
sysuse bpwide, clear
anova bp_after sex agegrp sex*agegrp, sequential
I'm trying to -test- based upon so-called SAS Type II sums of squares, here.
Now, I can manually calculate the Type II sums of squares from the -ereturn-
results for two -anova- runs (the difference between e(mss) from -anova depvar
A B, sequential- and that from -anova depvar A, sequential-) to do the test,
but it would seem more natural to have -test- acknowledge the options of the
-anova- command that it refers to . . .
* For searches and help try: