Event
Ting-Huei Chen, Universit茅 Laval
Thursday, May 4, 2017 15:30to16:30
Room D4-2019, Seminar Statistique Sherbrooke, 2500, boul. de l'Universit茅, Sherbrooke, CA
The Role of Tuning Parameters in High Dimensional Problems.
Various forms of penalty functions have been developed for regularized estimation. Screening approaches are often used to reduce the number of covariate before penalized estimation. However, in certain problems, the number of covariates remains large after screening. For example, in genome-wide association studies, the purpose is to identify Single Nucleotide Polymorphisms (SNPs) that are associated with certain traits. Because of the strong correlation of nearby SNPs, screening can only reduce the number of SNPs from millions to tens of thousands. Several penalty functions have been proposed for such high dimensional data. However, it is unclear which class of penalty functions is the appropriate choice for a particular application. In this talk, I will discuss the results of the theoretical analysis to relate the ranges of tuning parameters of various penalty functions with the dimensionality of the problem and the minimum effect size.