The Lorenz regression procedure introduced by [4] aims to estimate the explained Gini coefficient, a quantity with a natural application in the field of inequality of opportunity. In this paper, we introduce a lasso-type estimator for the explained Gini coefficient and discuss the selection of the regularization parameter. The performance of the procedure is compared to an oracle estimator on simulated data. Finally, an illustration on real-data is provided.
Jacquemain, A., Heuchenne, C., & Pircalabelu, E. (2021). A lasso-type estimation for the Lorenz regression (LIDAM Discussion Paper ISBA 2021/41). https://hdl.handle.net/2078.5/103396