B1251
Title: On Lasso regression for complex survey data: A new replicate weights cross-validation proposal
Authors: Amaia Iparragirre - University of the Basque Country (Spain)
Thomas Lumley - University of Auckland (United States)
Inmaculada Arostegui - University of The Basque Country (Spain)
Irantzu Barrio - University of the Basque Country (Spain) [presenting]
Abstract: LASSO regression models are one of the most commonly used variable selection methods, for which cross-validation is the most widely applied validation technique to choose the tuning parameter. Validation techniques in a complex survey framework are closely related to replicate weights. Applying LASSO regression models to complex survey data could be challenging. The goal is twofold: On the one hand, the performance of replicate weights methods is analyzed to select the tuning parameter for fitting LASSO regression models to complex survey data. On the other hand, new replicate weights methods are proposed for the same purpose. In particular, a new design-based cross-validation method is proposed as a combination of the traditional cross-validation and replicate weights. The performance of all these methods has been analyzed and compared using an extensive simulation study to the traditional cross-validation technique to select the tuning parameter for LASSO regression models. The results suggest a considerable improvement when the new proposal design-based cross-validation is used instead of the traditional cross-validation.
