About this question

Lasso Coordinate Descent

Hard · machine_learning · Quant Researcher interview question · lasso, coordinate_descent, regularization, feature_selection, machine_learning

Lasso regression is a key technique in quantitative finance for building sparse factor models, pruning hundreds of alpha signals into a parsimonious set. The coordinate descent algorithm provides an efficient, matrix-inversion-free method to solve the L1-penalized objective by iteratively updating each coefficient. This approach is fundamental to implementing scalable feature selection for return prediction and risk modeling. Task Implement the function lasso_coordinate_descent(X, y, alpha, max