This announcement is being sent on behalf of Deborah Balk.

-----------------------------------------------------

From: Lorna Thorpe [mailto:[log in to unmask]] 
Subject: SPRING 2013: Bayesian Statistics Course Advertisement

Students and Faculty,

See below for description of an exciting new course being offered this Spring by one of CUNY SPH’s newest faculty.
Please consider this course when registering!

Title: "Introduction to Bayesian Statistics for Translational Medicine: from Personal to Public Health" 
Th 6:05 – 8:00pm    Silberman Building (119th and 3rd Ave)

 Why take this course? 
The recent advent of translational medicine paradigm requires the merging of data analysis from molecular, individual, and community levels.  Modern technologies (e.g., online survey and genomic sequencing) make it popular to see data with high dimension (e.g., hundreds of genes), large sample size (e.g., thousands of Facebook users), or complex structure (e.g., patients nested in hospitals and then further nested in regions).  All these create urgent need for statistical methods that are powerful, flexible, integrative, and context-dependent.  Bayesian inference with Monte Carlo simulation offers such a promising solution! 

 What will be covered? 
(1) Introduction to principles of Bayesian statistical learning and elicitation of prior distributions; (2) Markov Chain Monte Carlo (MCMC) algorithms and computing software packages (e.g., WinBUGS); (3) Bayesian adaptive design in epidemiology and clinical studies; (4) Bayesian hierarchical modeling for longitudinal, multi-level, and spatial data; (5) applications to large-scale data with complex structured data (e.g., GWAS and nationwide survey data).

 Who should come?
To take credit, you should have taken introductory level biostatistics (e.g., PH 750).  Knowing some programming skills in R is a plus, but not necessary. You are welcome to audit and join us anytime.   

 Instructor:
Xiaowei Yang, Ph.D., Associate Professor at CUNY-School of Public Health, has many years of medical research with Bayesian inferences. As a NIH-funded investigator, he has developed MCMC algorithms for longitudinal data with missing values, integrative biomarker identification using Bayesian Variable Selection, and Bayesian adaptive design for personalized medicine development.  

<>

If you wish to unsubscribe from the CIDRANNOUNCEMENT-L List, please send an E-mail to:
"[log in to unmask]".  Within the body of the text, only write the following:
"SIGNOFF CIDRANNOUNCEMENT-L".