Sparse Portfolio Selection Via Non-convex Fraction Function
Angang Cui,Jigen Peng,Chengyi Zhang,Haiyang Li,Meng Wen
DOI: https://doi.org/10.48550/arxiv.1801.09171
2018-01-01
Abstract:In this paper, a continuous and non-convex promoting sparsity fraction function is studied in two sparse portfolio selection models with and without short-selling constraints. Firstly, we study the properties of the optimal solution to the problem (FP_a,λ,η) including the first-order and the second optimality condition and the lower and upper bound of the absolute value for its nonzero entries. Secondly, we develop the thresholding representation theory of the problem (FP_a,λ,η). Based on it, we prove the existence of the resolvent operator of gradient of P_a(x), calculate its analytic expression, and propose an iterative fraction penalty thresholding (IFPT) algorithm to solve the problem (FP_a,λ,η). Moreover, we also prove that the value of the regularization parameter λ>0 can not be chosen too large. Indeed, there exists λ̅>0 such that the optimal solution to the problem (FP_a,λ,η) is equal to zero for any λ>λ̅. At last, inspired by the thresholding representation theory of the problem (FP_a,λ,η), we propose an iterative nonnegative fraction penalty thresholding (INFPT) algorithm to solve the problem (FP_a,λ,η^≥). Empirical results show that our methods, for some proper a>0, perform effective in finding the sparse portfolio weights with and without short-selling constraints.