Feature screening algorithm for high dimensional data

Currently, feature screening is becoming an important topic in the fields of machine learning and high-dimensional data analysis. Filtering out irrelevant features from a set variables considered to be preliminary step that should performed before any Many approaches have been proposed same after work Fan Lv (J. Royal Stat. Soc., Ser. B. 70 (5), 849–911 (2008)), who introduced sure property. However, performance these methods differs one paper another. In this work, we aim add list new algorithm performing inspired by Kendall interaction filter Appl. 50 (7), 1496–1514 (2020)) when response variable continuous. The good behavior our proved through comparison with existing method, under several simulation scenarios.