Closed RJT1990 closed 8 years ago
Done - will be in next release. Now when a user fits a model, they will have a choice of showing transformed or untransformed results in the summary option. E.g:
model = ... x = model.fit('MLE') x.summary(transformed=False)
Will yield untransformed latent variables, with associated standard errors
For some models - e.g. GARCH - parameters are constrained so optimization is smoother, but it means when they are printed, the standard errors are not printed alongside. Should ideally untransform them after optimization is complete, then pass them through the likelihood to get the Inverse Hessian -> standard errors