It was really nice. About 40 people there. Including an
old friend from my NLP days who's now a prof at Michigan.
I mainly went through examples, showed the trailer halfway
through, showed the animations of MCMC, did a little work
on the board when people asked why Gibbs and Metropolis was
so slow. Then about 12 of us went to a great craft beer bar
right next to the venue.
I think I have my slide deck roughly where I want it, but
I need a couple more examples of how Stan/MCMC is used to compute
event probabilities, multiple comparisons, and estimates.
Next time.
- Bob
> On Aug 6, 2015, at 10:28 AM, Daniel Lee <
bea...@alum.mit.edu> wrote:
>
> Just curious how it went.
>
>
> Daniel
>
> --
> You received this message because you are subscribed to the Google Groups "stan development mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to
stan-dev+u...@googlegroups.com.
> For more options, visit
https://groups.google.com/d/optout.