Temos o prazer de convidar a todos para mais um seminário do Programa de Pós-Graduação em Estatística da UFRGS (PPGEst-UFRGS).
Título: Wavelet feature screening in high dimensional regression problems.
Resumo: In modern data analysis, the number p of explanatory variables is commonly much larger than the sample size n. Some classic methods to deal with this type of problem are sub-sampling covariates, ridge regression and Lasso (Friedman et al., 2009). Besides the high computational cost, p much larger than n may also affect the performance of variable selection methods. This motivates screening covariates.
Thence, an initial screening of relevant covariates is common in high-dimensional models. One important issue here is that many important features might have a relevant albeit highly nonlinear relation with the response. We propose a wavelet feature screening method that can be easily implemented, and we prove that, under suitable conditions, it captures the true covariates with high probability. Simulation results also show that our approach outperforms other screening methods in highly nonlinear models. We apply feature screening to two datasets about ozone concentration and epilepsy. In both applications, the proposed method selects features that match findings in the literature of their respective research fields, illustrating the applicability of feature screening.
This is a joint work with Rodney Fonseca and Pedro Morettin.
Mediador: Prof. Dr. Flavio Augusto Ziegelmann