CORRGAN: Sampling Realistic Financial Correlation Matrices Using Generative Adversarial Networks

Abstract

We propose a novel approach for sampling realistic financial correlation matrices. This approach is based on generative adversarial networks. Experiments demonstrate that generative adversarial networks are able to recover most of the known stylized facts about empirical correlation matrices estimated on asset returns. This is the first time such results are documented in the literature. Practical financial applications range from trading strategies enhancement to risk and portfolio stress testing. Such generative models can also help ground empirical finance deeper into science by allowing for falsifiability of statements and more objective comparison of empirical methods.

Matrix Evolutions: Synthetic Correlations and Explainable Machine Learning for Constructing Robust Investment Portfolios

“…A straightforward approach is to consider the matrix evaluations of empirical correlation matrixes. Marti (2019) described them as follows and referred to classical literature on network complexity and hierarchy in financial markets: § A distribution of pairwise correlations that is significantly shifted to the positive § Eigenvalues that follow the Marchenko-Pastur distribution but for…”

Interpretable Machine Learning for Diversified Portfolio Construction

“…The framework we introduced in this article would be a suitable testbed to challenge them against the classical HRP strategy from López de Prado. Moreover, the analysis can be enhanced by comparing other strategies or enriching the training dataset by generating more complex simulations using AI such as generative adversarial networks (see, e.g., Wiese et al 2019 andMarti 2019) or using the matrix evolutions scheme of Papenbrock et al (2021).…”

Adaptive Seriational Risk Parity and Other Extensions for Heuristic Portfolio Construction Using Machine Learning and Graph Theory

“…Recent approaches for simulating realistic financial correlation matrixes explicitly address hierarchy as stylized facts (see Huettner and Mai 2019;Marti 2019;Jaeger et al 2021;Papenbrock et al 2021). Modeling the correlation hierarchy of markets has also been used for the recognition of market regimes (Papenbrock and Schwendner 2015).…”