Mesoscopic Structure of the Stock Market and Portfolio Optimization

TL;DR

This paper investigates the mesoscopic structure of the stock market using clustering techniques from Random Matrix Theory, revealing a stable intermediate structure that improves portfolio optimization and risk prediction.

Contribution

It introduces a novel mesoscopic market structure analysis, leveraging RMT-based clustering to enhance portfolio construction and propose a new wealth allocation scheme.

Findings

Filtering out random and systemic co-movements improves risk prediction.

Cluster-based portfolios show increased reliability and stability.

The proposed wealth allocation scheme enhances portfolio performance.

Abstract

The idiosyncratic (microscopic) and systemic (macroscopic) components of market structure have been shown to be responsible for the departure of the optimal mean-variance allocation from the heuristic `equally-weighted' portfolio. In this paper, we exploit clustering techniques derived from Random Matrix Theory (RMT) to study a third, intermediate (mesoscopic) market structure that turns out to be the most stable over time and provides important practical insights from a portfolio management perspective. First, we illustrate the benefits, in terms of predicted and realized risk profiles, of constructing portfolios by filtering out both random and systemic co-movements from the correlation matrix. Second, we redefine the portfolio optimization problem in terms of stock clusters that emerge after filtering. Finally, we propose a new wealth allocation scheme that attaches equal importance to stocks belonging to the same community and show that it further increases the reliability of the constructed portfolios. Results are robust across different time spans, cross-sectional dimensions and set of constraints defining the optimization problem