Stan meetup tonight in NYC: How Stan computes the posterior distribution


#1

Hey all,

We’re running another Stan meetup in NYC tonight. Sorry for the short notice.

I’ll be talking about how Stan computes lp__. It starts at 7 pm.

Register here: https://www.meetup.com/bda-group/events/250819101/


#2

These meetups sound great, but I’m busy most weeknights.

P.S. lp__ is what we used to call the target log density in the language and what it’s still called in the interface output. Now it’s target in the language.


#3

It’s target in the language, but lp__ in the output.

I broke down the lp__ into the part that’s the target in the model and then the log of the absolute value of the determinant of the Jacobian of the inverse transformation. I’ll post the slides soon.


#4

I put up links to the slides and the Stan programs up on my blog: http://syclik.com/2018/06/19/meetup-slides/.