Are the Classical Variable Selection Techniques Obsolete?

• Activity Number: 167
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): F. Michael Speed*+
• Companies: SAS
• Keywords: data mining ; LASSO ; variable selection

Abstract

In data mining most analysts use the classical forward, backward stepwise, or "all-possile subsets" techniques. GLMSELECT procedure performs variable selection in the framework of general linear models. A variety of model selection methods are available, including the LASSO method of Tibshirani (1996) and the related LAR method of Efron et al. (2004). The procedure offers extensive capabilities for customizing the selection with a wide variety of selection and stopping criteria, from traditional and computationally efficient significance-level-based criteria to more computationally intensive validation-based criteria. The procedure also provides graphical summaries of the selection search. The GLMSELECT procedure compares most closely to REG and GLM. The REG procedure supports a variety of model-selection methods but does not support a CLASS statement. The GLM procedure supports a CLASS statement but does not include effect selection methods. The GLMSELECT procedure fills this gap. This talk will focus on the selection techniques as well as the bootstrapping features. (Modified from SAS Help). I think that the answer to the question is YES!