Study on the performance of Robust LASSO in determining important variables data with outliers

A variable selection method is required to deal with regression models many variables, and LASSO has been the most widely used methodology. However, as several authors have noted, sensitive outliers in data. For this reason, Robust-LASSO approach was introduced by applying some weighting schemes for each sample This research presented a comparative study of three Robust LASSO, namely Huber-LASSO, Tukey-LASSO, Welsch-LASSO. The did rich simulation containing scenarios various characteristics on covariance structures explanatory variable, types outliers, number location active variables. then found that Tukey-LASSO outperformed Huber-LASSO Welsch-LASSO identifying significant performance generally decreased covariances among variables increased data dimension increased. Exploration sembung leaf extract shows high dimensional which contains about 14,28% response 25,71% Based research, selected using nine compounds, were eight 13 compounds. prediction accuracy superior other methods.