Beyond Co-Movement: Locality by Exposures Enables a Joint Factor-Graph Framework for Portfolio Diversification

2026-08-06

AI Summary

This study addresses the limitations of existing portfolio construction methods, which often fail to effectively integrate the complementary information from factor models and graph structures—either neglecting idiosyncratic shocks or overlooking the latent data structure underlying systematic returns. To overcome this, we propose the MINGLE framework, which innovatively constructs a graph based on the similarity of assets’ factor exposures, thereby aligning more closely with economic sectors. Within a unified ADMM optimization framework, MINGLE jointly learns factor representations and graph topology, enabling mutual regularization between the factor and graph domains. Empirical results demonstrate that our approach significantly outperforms correlation-based benchmarks across varying levels of volatility and transaction costs, with statistical tests confirming that the performance gains stem from the effective fusion of factor and graph information.

Abstract

Current portfolio construction methods are either agnostic to the effects of idiosyncratic shocks (standard factor models) or to the latent data structure driving systematic returns (recent graph-based approaches). This presents an opportunity to combine the complementary market aspects captured by the factor and graph domains, allowing asset allocations to operate directly on the underlying market structure, rather than on its observed co-movement or its finite-sample artefacts. In this work, we introduce the Mutually-INformed Graph-Locality and Exposures framework (MINGLE), which mutually regularises the factor and graph domains by redefining graph locality through systematic factor exposure profiles, rather than via observed co-movements. This is formalised through a unified Alternating Direction Method of Multipliers (ADMM) framework that jointly learns a latent factor representation and its induced graph topology directly from market returns. The resulting exposure-similarity graph aligns more closely with established economic sectors than conventional correlation-based graphs. Portfolios constructed from this representation are shown to consistently outperform their correlation-based counterparts across a range of volatility regimes and transaction cost levels. For rigour, paired statistical testing confirms that these gains stem from the reconciliation of the graph and factor domains.

Problem

• portfolio diversification
• factor models
• graph-based approaches
• systematic returns
• idiosyncratic shocks

Innovation

• factor-graph integration
• exposure-based locality
• portfolio diversification
• ADMM optimization
• latent graph learning