LINselect
LINselect is a R package which allows to estimate the mean of a Gaussian vector, by choosing among a large collection of estimators
LINselect is a R package which allows to estimate the mean of a Gaussian vector, by choosing among a large collection of estimators. In particular it solves the problem of variable selection by choosing the best predictor among predictors emanating from different methods as lasso, elastic-net, adaptive lasso, pls, randomForest. Moreover, it can be applied for choosing the tuning parameter in a Gauss-lasso procedure.
Il y a une 2ème publication de référence : http://dx.doi.org/10.1214/13-AIHP539
LINselect est aussi disponible sur le site du CRAN : https://cran.r-project.org/web/packages/LINselect/