Kevin Sheppard
Kevin Sheppard
Yes, it assumes no correlation which is the usual practice for simple statistics. Improved documentation is always welcome. Note that there was a bug fix in #8227, that will be...
> AIC is a relative measure used for model selection - and for feature selection in particular. It's confusing that AIC can be reported by a single model - it...
Possibly. I haven't started thinking about Dynamic Panel models to think about what a good interface would look like. I don't think the univariate GMM IV is enough to estimate...
A gist or similar would be helpful.
Nothing yet, but doesn't seem too hard to implement.
https://www.ssc.wisc.edu/~bhansen/papers/ecnmt_21.pdf
A PR would be welcome.
One option would be to add support where it is possible, e.g., IVGMM, but not for models that are not supported.
Inequality constraints are not very standard, and inference, when constraints bound, is not simple. You could always do model selection to remove negative coefficients.
Effects are errors and so are not part of the prediction. They are not consistently estimated in the classic fixed T, large N setting. If one dimension is fixed and...