Interpreting Linear Beta Coefficients Alongside Feature Importances in Machine Learning
Sign Up to like & getrecommendations! Published in 2021 at "Atlantic Economic Journal"
DOI: 10.1007/s11293-021-09711-8
Abstract: Machine-learning regression models lack the interpretability of their conventional linear counterparts. Tree- and forest-based models offer feature importances, a vector of probabilities indicating the impact of each predictive variable on a model’s results. This brief… read more here.
Keywords: beta coefficients; machine learning; feature importances; importances machine ... See more keywords