Event space & lunch sponsored by Viacom
Stan (http://mc-stan.org) is a statistical modeling platform used by thousands of scientists, engineers, & other researchers for statistical modeling, data analysis, & prediction. It is being applied academically & commercially across fields as diverse as ecology, pharmacometrics, physics, political science, finance & econometrics, professional sports, real estate, publishing, recommender systems, & educational testing.
In this workshop well review the foundations of Bayesian inference & computation, playing specific emphasis on the details critical to robust statistical analyses. Well then demonstrate the implementation of these methods in Stan with interactive examples, beginning with parametric regression & classification before considering their Gaussian process equivalents.
Speaker Bios:Michael Betancourt is a research scientist in the Applied Statistics Center at Columbia University, where he develops theoretical & methodological tools to support practical Bayesian inference. He is also a core developer of Stan, where he implements & tests these tools. In addition to hosting tutorials & workshops on Bayesian inference with Stan he also collaborates on analyses in, amongst others, epidemiology, pharmacology, & physics.
Mitzi Morris is a member of the Stan development team. Her background is in naturallanguage processing & bioinformatics. Her editor is emacs.
10:00 - 11:00Foundations of Bayesian Inference & Computation
11:00 - 11:30Introduction to Stan
11:30 - 12:00Linear Regression
12:00 - 01:00Lunch
01:00 - 01:30Linear Regression (cont)
01:30 - 02:30Logistic Regression
02:30 - 03:00Introduction to Gaussian Processes
03:00 - 04:00Gaussian Process Regression
04:00 - 05:00Gaussian Process Classification
Most of the course will be interactive examples so the schedule will adapt to the students speed.
Requirements:PyStan 2.16.0,http://pystan.readthedocs.io/en/latest/matplotlibDownload Stan 2.16.0 Manual,1.79 MB stan-reference-2.16.0.pdf