Past Meetup

Bayesian Generalized Linear Models and Appropriate Default Prior

This Meetup is past

188 people went

Location visible to members

Details

For our first event of 2012, we are excited to have Dr. Andrew Gelman, Professor of Statistics and Political Science at Columbia University, present his work to the group.

This event will be sponsored by Revolution Analytics who will be providing pizza AND will be raffling off a full licence of Revolution R Enterprise version 5.0!! So come early to join the fun! The license will be for 6 months and will be extended to a year if the winner participates in a short phone survey after 6 months.

Special thanks goes to Jared Lander for not only introducing us to Prof. Gelman, but also for setting up the sponsorship for the event. Thanks Jared!

Pizza / networking will begin at 6:15pm. The presentation will begin at 7pm.

Abstract:

Many statistical methods of all sorts have tuning parameters. How can default settings for such parameters be chosen in a general-purpose computing environment such as R? We consider the example of prior distributions for logistic regression. Logistic regression is an important statistical method in its own right and also is commonly used as a tool for classification and imputation. The standard implementation of logistic regression in R, glm(), uses maximum likelihood and breaks down under separation, a problem that occurs often enough in practice to be a serious concern. Bayesian methods can be used to regularize (stabilize) the estimates, but then the user must choose a prior distribution. We illustrate a new idea,t he "weakly informative prior," and implement it in bayesglm(), a slight alteration of the existing R function. We also perform a cross-validation to compare the performance of different priordistributions using a corpus of data sets.

Bio:

Andrew Gelman is a professor of statistics (http://www.stat.columbia.edu/) and political science (http://www.columbia.edu/cu/polisci/) and director of the Applied Statistics Center (http://applied.stat.columbia.edu/) at Columbia University. He has received the Outstanding Statistical Application award from the American Statistical Association, the award for best article published in the American Political Science Review, and the Council of Presidents of Statistical Societies award for outstanding contributions by a person under the age of 40. His books include Bayesian Data Analysis (http://www.stat.columbia.edu/~gelman/book/) (with John Carlin, Hal Stern, and Don Rubin), Teaching Statistics: A Bag of Tricks (http://www.stat.columbia.edu/~gelman/bag-of-tricks) (with Deb Nolan), Data Analysis Using Regression and Multilevel/Hierarchical Models (http://www.stat.columbia.edu/~gelman/arm/) (with Jennifer Hill), and, most recently,Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do (http://redbluerichpoor.com/) (with David Park, Boris Shor, Joe Bafumi, and Jeronimo Cortina).

Andrew has done research on a wide range of topics, including: why it is rational to vote; why campaign polls are so variable when elections are so predictable; why redistricting is good for democracy; reversals of death sentences; police stops in New York City, the statistical challenges of estimating small effects; the probability that your vote will be decisive; seats and votes in Congress; social network structure; arsenic in Bangladesh; radon in your basement; toxicology; medical imaging; and methods in surveys, experimental design, statistical inference, computation, and graphics.

We hope to see you there!