February 2012
Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29      

Authors' Committee


Matt Blackwell (Gov)


Martin Andersen (HealthPol)
Kevin Bartz (Stats)
Deirdre Bloome (Social Policy)
John Graves (HealthPol)
Rich Nielsen (Gov)
Maya Sen (Gov)
Gary King (Gov)

Weekly Research Workshop Sponsors

Alberto Abadie, Lee Fleming, Adam Glynn, Guido Imbens, Gary King, Arthur Spirling, Jamie Robins, Don Rubin, Chris Winship

Weekly Workshop Schedule

Recent Comments

Recent Entries



SMR Blog
Brad DeLong
Cognitive Daily
Complexity & Social Networks
Developing Intelligence
The Education Wonks
Empirical Legal Studies
Free Exchange
Health Care Economist
Junk Charts
Language Log
Law & Econ Prof Blog
Machine Learning (Theory)
Marginal Revolution
Mixing Memory
Mystery Pollster
New Economist
Political Arithmetik
Political Science Methods
Pure Pedantry
Science & Law Blog
Simon Jackman
Social Science++
Statistical modeling, causal inference, and social science



Powered by
Movable Type 4.24-en

« App Stats: Sofer on "Sparse Joint Estimation of Covariates-Dependent Covariance Matrices" | Main | App Stats: Pfister on "Visual Computing in Biology" »

20 February 2012

App Stats: Dominici on "Bayesian Effect Estimation Accounting for Adjustment Uncertainty"

We hope you can join us this Wednesday, February 22, 2012 for the Applied Statistics Workshop. Francesca Dominici, Professor of Biostatistics from the Department of Biostatistics at the Harvard School of Public Health, will give a presentation entitled "Bayesian Effect Estimation Accounting for Adjustment Uncertainty". A light lunch will be served at 12 pm and the talk will begin at 12.15.

"Bayesian Effect Estimation Accounting for Adjustment Uncertainty"
Francesca Dominici
Department of Biostatistics, Harvard School of Public Health
CGIS K354 (1737 Cambridge St.)
Wednesday, February 22nd, 2012 12.00 pm


Model-based estimation of the effect of an exposure on an outcome is generally sensitive to the choice of which confounding factors are included in the model. We propose a new approach, which we call Bayesian Adjustment for Confounding (BAC), to estimate the effect on the outcome associated with an exposure of interest while accounting for the uncertainty in the confounding adjustment. Our approach is based on specifying two models: 1) the outcome as a function of the exposure and the potential confounders (the outcome model); and 2) the exposure as a function of the potential confounders (the exposure model). We consider Bayesian variable selection on both models and link the two by introducing a dependence parameter ω denoting the prior odds of including a predictor in the outcome model, given that the same predictor is in the exposure model. In the absence of dependence (ω = 1), BAC reduces to traditional Bayesian Model Averaging (BMA). In simulation studies we show that BAC with ω > 1 estimates the exposure effect with smaller bias than traditional BMA, and improved coverage. We then compare BAC, a recent approach of Crainiceanu et al. (2008), and traditional BMA in a time series data set of hospital admissions, air pollution levels and weather variables in Nassau, NY for the period 1999-2005. Using each approach, we estimate the short-term effects of PM2.5 on emergency admissions for cardiovascular diseases, accounting for confounding. This application illustrates the potentially significant pitfalls of misusing variable selection methods in the context of adjustment uncertainty.

Posted by Konstantin Kashin at February 20, 2012 2:03 AM