Abstract | ||
---|---|---|
Maximum likelihood estimation (MLE) is a fundamental computational problem in statistics. In this paper, MLE for statistical models with discrete data is studied from an algebraic statistics viewpoint. A reformulation of the MLE problem in terms of dual varieties and conormal varieties will be given. With this description, we define the dual likelihood equations and the dual MLE problem. We show that solving the dual MLE problem yields solutions to the MLE problem, and that we can solve the dual MLE problem even if we do not have the defining equations of the model itself. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1145/2631948.2631959 | SNC |
Keywords | Field | DocType |
statistical computing,maximum likelihood estimation,algebraic statistics,critical points,dual varieties | Computational problem,Maximum likelihood,Statistical model,Statistics,Algebraic statistics,Defining equation (physics),Mathematics | Conference |
Citations | PageRank | References |
4 | 0.55 | 4 |
Authors | ||
1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jose Israel Rodriguez | 1 | 17 | 6.01 |