[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: error message about Mata memory
In response to my question about the details of his model, Jeph Herrin
> Thanks for the informative response. I guess "simple" could mean many
> things, here I only meant there were no random coefficients. There are
> however, 50 covariates.
> From this I deduce that if I can reduce my set of covariates substantially,
> I might be able to estimate something close to what I want.
Of course! One random intercept, but 50 covariates (i.e. 50 betas) would
do it. It was dense of me not to realize that.
* For searches and help try: