non-standardized T distribution

[atteson]

It would be nice to have a non-standardized t distribution:

http://en.wikipedia.org/wiki/Student's_t-distribution#Non-standardized_Student.27s_t-distribution

It is a straightforward generalization of the t distribution. I realize this is a "feature request" rather than an "issue". Please let me know if there's some other way that I'm supposed to post this.

[johnmyleswhite]

GitHub doesn't really let us distinguish feature requests from issues, except via tags, so this is the right place.

The best possible way to push this forward is to try implementing the distribution and submitting a pull request. Once you've got something that seems to be functional, people will be in a much better position to comment on your work.

[simonbyrne]

Actually, there is one, it's just not documented yet:

julia> using Distributions

julia> NoncentralT(3,1.5)
NoncentralT( df=3.0 ncp=1.5 )

[simonbyrne]

Sorry, your issue title doesn't match your question! It should be fairly straightforward to implement (look at other location-scale families, e.g. Logistic or Cauchy).

On the other hand, perhaps we would be better off adding the UnivariateLocationScaleFamily from #144 (even if we don't support all the operations yet)?

[atteson]

Oops. Thanks for pointing that out. I meant non-standardized t.
Non-central t is something else.

On Thu, Sep 25, 2014 at 6:32 AM, Simon Byrne notifications@github.com
wrote:

Sorry, your issue title doesn't match your question! It should be fairly
straightforward to implement (look at other location-scale families (e.g.
Logistic or Cauchy).

On the other hand, perhaps we would be better off adding the
UnivariateLocationScaleFamily from #144
#144 (even if we
don't support all the operations yet)?

—
Reply to this email directly or view it on GitHub
#286 (comment)
.

[johnmyleswhite]

Pulling the core conceptual machinery out of #144 would be a great idea. I'll try to do that over the weekend.

[lindahua]

I change the title to make it accurately reflect what it really intends.

[chipkent]

I just submitted a pull request with this feature implemented. Pull-490