A Variety of Introductory Topics from Algebraic Statistics

James McVittie

12:00, Friday, Jan. 31
BURN 1025



In parametrically defined statistical models, maximum likelihood estimators may be used to obtain estimators for the unknown parameters. The procedure of finding the arguments which maximize the likelihood is equivalent to solving a set of score equations (i.e. partial derivatives of the log likelihood set to 0). When the statistical models are defined in such a way that the score functions are polynomials, the resulting maximum likelihood estimators correspond to the algebraic variety of a set of equations.

In this talk, we will attempt to link some ideas from the world of statistical inference to ideas in the world of algebraic geometry. We will examine a sequence of nucleobases, and by proposing a parametric statistical model for the way in which the nucleobases were selected, we will estimate the unknown paramaters using a Gröbner basis.

All graduate students are invited. As with all talks in the graduate student seminar, this talk will be accessible to all graduate students in math and stats.

This seminar was made possible by funding from the McGill mathematics and statistics department and PGSS.

back