[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: re: xtmixed question
When I use a maximum likelihood mixed model with clusters that have
unequal sizes, are the fixed effects modeled in a way that reflects
The estimation method does deal indirectly with unequal sample sizes
I have been told: "larger groups impose larger influences on the
likelihood function to be maximized..."
So in a model like,
xtmixed y treatment || group:
where multiple groups are each assigned to one of two treatments,
but groups are different sizes, is this accounted for by these
types of models?
So if I were to average the groups and use the summary statistics
in a linear regression model, I might weight the analysis by
cluster size, reducing influence of small clusters relative to
large clusters. If I use xtmixed directly, should I worry about this?
* For searches and help try: