How to get confidence distribution using MC??

[Anand]

Hi all

I am trying to understand how to get confidence distribution using
monte carlo simulation based on Joel blog

http://www.joelonsoftware.com/items/2007/10/26.html

suppose i get the sum of esimates/Random velocity as 71.3

I repeat steps 100 times & get the 100 values like {71.3, 96, 63,
75..............}
Similarly i get the data for all the other developers in the team.

What are the api's to create cumulative distribution function (CDF) to
get confidence distribution graph? How do i do that using pymc?

-Thanks & Regards
Anand

[Anand Patil]

Hello self,

Very interesting webpage. You could do something similar, but a bit
more sophisticated using a Bayesian hierarchical model in PyMC. I'd
suggest checking out Gelman's _Bayesian Analysis_.

If you want to replicate their empirical method exactly, however, PyMC
is overkill... you can easily draw random elements from a set, produce
empirical CDF's etc. using NumPy.

Have fun,
Anand

> --
>
> You received this message because you are subscribed to the Google Groups "PyMC" group.
> To post to this group, send email to py...@googlegroups.com.
> To unsubscribe from this group, send email to pymc+uns...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/pymc?hl=en.
>
>
>