Nonlinear variable selection with continuous outcome: a nonparametric incremental forward stagewise approach
This is an incremental method for statisticians and data scientists needing flexible variable selection in high-dimensional settings.
The authors tackled variable selection in sparse generalized additive models without assuming a functional form, using a nonparametric incremental forward stagewise approach with roughening for residual adjustment, and showed it is competitive against popular machine learning methods in simulations and real datasets.
We present a method of variable selection for the sparse generalized additive model. The method doesn't assume any specific functional form, and can select from a large number of candidates. It takes the form of incremental forward stagewise regression. Given no functional form is assumed, we devised an approach termed roughening to adjust the residuals in the iterations. In simulations, we show the new method is competitive against popular machine learning approaches. We also demonstrate its performance using some real datasets. The method is available as a part of the nlnet package on CRAN https://cran.r-project.org/package=nlnet.