Causality and Statistics
In many statistics classes, you'll hear or say "correlation doesn't equal causation." While this is true, statistical tools and techniques can be useful for exploring causal claims, even in observational studies. In this talk, I'll give an overview of some of these techniques, many coming from graph theory, to talk about when you should and shouldn't include covariates in models.
Additionally, I'll talk about what the job of a statistical consultant is like, with plenty of time for questions and discussion.
Matt Thomas, Statistical Consultant, Cornell Statistical, Consulting Unit, Cornell University
Monday, March 27
4pm in Williams 320
Lite refreshments will be served
sponsored by the Math Department Colloquium series
contact: tgalanthay@ithaca.edu (Prof. Ted Galanthay)