Sparse Portfolio Selection via Non-convex Fraction Function

TL;DR

This paper introduces a novel non-convex fraction function for sparse portfolio selection, develops theoretical properties and algorithms, and demonstrates effective empirical performance in finding sparse portfolios with or without short-selling constraints.

Contribution

The paper proposes a new non-convex fraction function, analyzes its properties, and develops iterative thresholding algorithms for sparse portfolio selection.

Findings

Effective in selecting sparse portfolios

Handles constraints with and without short-selling

Provides theoretical guarantees for solutions

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 $(FP_{a,\lambda,\eta})$ 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 $(FP_{a,\lambda,\eta})$. 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 $(FP_{a,\lambda,\eta})$. Moreover, we also prove that the value of the regularization parameter $\lambda>0$ can not be chosen too large. Indeed, there exists $\bar{\lambda}>0$ such that the optimal solution to the problem $(FP_{a,\lambda,\eta})$ is equal to zero for any $\lambda>\bar{\lambda}$. At last, inspired by the thresholding representation theory of the problem $(FP_{a,\lambda,\eta})$, we propose an iterative nonnegative fraction penalty thresholding (INFPT) algorithm to solve the problem $(FP_{a,\lambda,\eta}^{\geq})$. 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.