Smart network based portfolios

TL;DR

This paper enhances portfolio allocation methods by integrating network theory to improve risk estimation and portfolio performance, demonstrating superior out-of-sample results over traditional approaches.

Contribution

It introduces a network-based framework for portfolio optimization using different covariance estimators, showing improved risk management and performance.

Findings

Network-based portfolios outperform standard portfolios in risk and return.

Use of shrinkage covariance estimator improves portfolio stability.

Network approach reduces portfolio risk in high-dimensional settings.

Abstract

In this article we deal with the problem of portfolio allocation by enhancing network theory tools. We use the dependence structure of the correlations network in constructing some well-known risk-based models in which the estimation of correlation matrix is a building block in the portfolio optimization. We formulate and solve all these portfolio allocation problems using both the standard approach and the network-based approach. Moreover, in constructing the network-based portfolios we propose the use of two different estimators for the covariance matrix: the sample estimator and the shrinkage toward constant correlation one. All the strategies under analysis are implemented on two high-dimensional portfolios having different characteristics, covering the period from January $2001$ to December $2017$. We find that the network-based portfolio consistently better performs and has lower risk compared to the corresponding standard portfolio in an out-of-sample perspective.