Portfolio optimization with transfer entropy constraints

This paper integrates transfer entropy (TE) within the portfolio optimization framework to account for dependencies among assets. This approach helps mitigate systemic risk and create portfolios that are resilient to asymmetric information flows. Key contributions of this study include (1) demonstrating the impact of TE constraints on portfolio diversification and stability, (2) linking TE thresholds to the Herfindahl–Hirschman Index to quantify this effect, and (3) establishing the coherence of a TE-integrated multivariate entropic risk measure using extreme value theory. Empirical analyses of a diversified portfolio, including traditional and contemporary asset classes, reveal that TE constraints effectively modulate portfolio stability and offer a robust alternative to conventional risk measures such as Value at Risk and Conditional Value at Risk.