Hello group!
This is an research talk for undergraduates which may nevertheless be of
interest to several members of the group. I have heard nothing but
great things about Professor Blitzstein's abilities as a lecturer!
Unfortunately I will be out of town for this, but I highly recommend it.
Cheers,
Jacob
<-----Original Message----->
From: Preya Shah [shah(a)college.harvard.edu]
Sent: 3/26/2011 4:40:50 PM
To: quincy-open(a)lists.hcs.harvard.edu; sps-open(a)lists.hcs.harvard.edu;
thurj-staff(a)googlegroups.com
Subject: [sps-open] Cool Statistics Talk (and mochi ice cream!) this
Tuesdayat 8pm
Interested in models for finance, computational physics, or
bioinformatics?
Statistics?
Come hear
"The Statistical Magic of Markov Chain Monte Carlo"
with
Professors Xiao-Li Meng and Joseph Blitzstein
When: Tuesday, March 29 at 8 PM
Where: Sever 102
Markov chain Monte Carlo (MCMC) methods, which originated in
computational physics about a half century ago, have seen an enormous
range of applications in both the natural and social sciences, and
beyond. This is mainly due to their ability to simulate from very
complex distributions needed by all kinds of statistical models, from
bioinformatics to financial engineering to astronomy. The first part of
the talk provides an introductory tutorial for the two most frequently
used MCMC algorithms: the Gibbs sampler and the Metropolis-Hastings
algorithm. Using simple yet non-trivial examples, we demonstrate, via
live performance, the good, bad, and ugly implementations. The second
part of the talk shows some examples of how these algorithms are applied
in practice, enabling investigations which would have been considered
unimaginably complex before the advent of fast computers together with
MCMC techniques.
This talk will be an introduction to MCMC statistical models.
Come early to get ice cream mochi before it melts!
Sponsored by The Harvard Undergraduate Research Journal (THURJ)
_______________________________________________
sps-open mailing list
sps-open(a)lists.hcs.harvard.edu
https://lists.hcs.harvard.edu/mailman/listinfo/sps-open