Hello gretl users,
I am working with multiple imputation and I think it would be great to make
gretl work with it. The first issue is that I couldn't find any Gibbs
Sampler for gretl, so I have a couple of questions about how to solve that.
1.- How difficult would it be to program everything in gretl? I mean gibbs
sampling plus Rubin's inference.
2.- R can handle multiple impution pretty well. Would it be possible to run
multiple imputation within R and gretl just reads the results? Many MI
programs run R in the background and they work fine. Another way to do that
would be translate the code of R libraries to gretl.
Any idea or suggestion is pretty welcome.
Best regards,
Sebastian