Skip to content

Conversation

@calbertsen
Copy link
Contributor

Added class to evaluate a multivariate t distribution. It has been tested with the mvtnorm package in R.

@skaug
Copy link
Contributor

skaug commented Sep 8, 2014

Thanks, the multivariate t-distribution is definitely useful, but I wonder if it instead
belongs in "distributions_R.hpp"? E.g. does SEPARABLE(MVT(Sigma,Type(4)),MVT(Sigma,Type(4)) work? I expect that the whole machinery of "density.hpp" requires normality.

If I am right it should instead be called "mvtnorm".

Hans

@calbertsen
Copy link
Contributor Author

Hi Hans

It is true that the current implementations of SEPARABLE_t and PROJ_t requires normality, while the others do not (AR1_t, SCALE_t and VECSCALE_t).
So you are right that it does not work with SEPARABLE_t and PROJ_t, while it does work with the three others.
It should however be possible to implement versions of SEPARABLE_t and PROJ_t for the multivariate t-distribution.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants