RPS: Portfolio Asset Selection using Graph based Representation Learning

TL;DR

This paper introduces RPS, a novel method that uses representation learning and clustering to improve portfolio diversification and enhance existing optimization algorithms' performance.

Contribution

The paper presents RPS, a new approach that redefines asset distances with representation learning and clustering to select better asset subsets for portfolio optimization.

Findings

RPS improves diversification in portfolio selection.

Existing algorithms like MVO, CLA, and HRP benefit from RPS.

Empirical results show enhanced performance of portfolio optimization methods.

Abstract

Portfolio optimization is one of the essential fields of focus in finance. There has been an increasing demand for novel computational methods in this area to compute portfolios with better returns and lower risks in recent years. We present a novel computational method called Representation Portfolio Selection (RPS) by redefining the distance matrix of financial assets using Representation Learning and Clustering algorithms for portfolio selection to increase diversification. RPS proposes a heuristic for getting closer to the optimal subset of assets. Using empirical results in this paper, we demonstrate that widely used portfolio optimization algorithms, such as MVO, CLA, and HRP, can benefit from our asset subset selection.