Constrained Portfolio Optimization with Estimation Risk: Theory and ESG Applications

Zhenzhen Huang, Pengyu Wei, and Chengguo Weng

♦ This paper studies portfolio combination for mitigating estimation risk in the first two moments of asset returns under a constrained framework. We develop a combined three-fund portfolio and derive optimal combination coefficients by maximizing expected out-of-sample mean-variance utility under expected equality or inequality constraints. The proposed sophisticated estimators allow the portfolio to attain the target factor exposure pointwise. Using environmental, social, and governance (ESG) investing as an illustration in simulation and empirical studies, we demonstrate that our combined portfolios consistently outperform standard benchmarks by delivering higher expected out-of-sample utility while reliably anchoring to the target factor exposure.

Read it here.