Organizer: Professor Linda Chen (2-9947, MW 412, lchen@math.ohio-state.edu)
Algebraic Statistics is concerned with problems that lie at the intersection of algebra, geometry, combinatorics, and statistics. Methods from algebra and geometry can be used to make statistical inferences; many statistical models for discrete random variables can be represented by classical algebraic varieties, e.g. secant varieties and toric varieties.
The goal of this seminar will be to understand this connection and its statistical consequences, for example, in maximum likelihood estimation. We will also discuss applications to computational biology, in particular to genome sequence analysis. Further topics will be determined by the interests of the participants.
One of the goals of this seminar is to foster interactions between the Mathematical Biosciences Institute (MBI) and the Department of Mathematics. Graduate students, postdocs, and faculty members of all fields are welcome. No expertise in algebraic geometry, statistics, or math biology will be assumed.
| Date | Speaker | Title |
| September 25 | Linda Chen | Introduction and Overview |
| October 2 | Laura Kubatko | Statistics |
| October 9 | Laura Kubatko | Linear and Toric Models | October 16 | Discussion | Markov Models |
| October 23 | Discussion | Tree and Graphical Models |
| October 30 | Brandy Stigler | Log-Linear Models, Toric Varieties, Markov Bases |
| November 6 | Seth Sullivant (Harvard)
MBI Seminar Series 1:30-2:30pm, Jennings 355 |
Algebraic statistical models |
| November 6 | Seth Sullivant (Harvard)
Algebraic Geometry Seminar 4:30-5:30pm, Scott Lab 241 |
Algebraic geometry of Gaussian Bayesian networks |
| November 13 | Log-Linear Models, Toric Varieties, Markov Bases II | |
| November 20 | Kevin Woods (Oberlin) | Parametric Inference for Graphical Models |
| November 27 | Dennis Pearl |