Skip to content

Bayesian estimation with Markov Chain Monte Carlo using PyMC

Photo of Eric Novik
Hosted By
Eric N. and Drew C.
Bayesian estimation with Markov Chain Monte Carlo using PyMC

Details

Thanks to the New York Marathon (https://www.nycmarathon.org/), we are very lucky to have Chris Fonnesbeck (https://biostat.mc.vanderbilt.edu/wiki/Main/ChrisFonnesbeck), instructor in the Department of Biostatistics at Vanderbilt University, in town in early November. Chris has been kind enough to offer to present his work on Bayesian estimation using MCMC, despite the fact that he will no doubt be exhausted from having run a marathon the day before.

Chris is one of the co-creators of PyMC (https://code.google.com/p/pymc/), the premier package for MCMC estimation in Python. From PyMC's website:

Bayesian estimation, particularly using Markov chain Monte Carlo (MCMC), is an increasingly relevant approach to statistical estimation. However, few statistical software packages implement MCMC samplers, and they are non-trivial to code by hand. PyMC is a python module that implements the Metropolis-Hastings algorithm as a python class, and is extremely flexible and applicable to a large suite of problems. PyMC includes methods for summarizing output, plotting, goodness-of-fit and convergence diagnostics.

You'll note that this presentation does not directly involve R, and there are a couple of reasons for this:

Chris has greatly contributed to the statistical computing community, and we are lucky to have him
Given the recent consternation within the R community (https://www.stat.columbia.edu/~cook/movabletype/archives/2010/09/ross_ihaka_to_r.html) about the language, it is valuable to understand how other languages approach statistical computing
We always say the R meetup is the satellite Python meetup, and now we can make that official. I have also invited the Python users group (https://www.meetup.com/nycpython/) to this meetup, and I hope it will be an opportunity for these communities to mingle.

Chris is going to send me a brief description of his talk in the coming days, and I will update this announcement when I receive it.

Finally, depending on time and interest level, I may also present a brief follow-up to Chris's talk on integrating R with PyMC using some of the various tools that exist to bridge R to Python.

Please let me know if you have any questions.

Photo of New York Open Statistical Programming Meetup group
New York Open Statistical Programming Meetup
See more events
Interschool Lab
Columbia University, CEPSR Hall, Room 750 · New York, NY