A1341
Title: Multiobjective ESG bond portfolio optimization
Authors: Maziar Sahamkhadam - Linnaeus University (Sweden) [presenting]
Andreas Stephan - Linnaeus University (Sweden)
Abstract: The purpose is to develop a copula-based pricing model for forecasting bond returns and to apply it to multi-objective bond portfolio (MOBP) optimization. Evidence of asymmetric tail dependence is provided in the zero-coupon bond yield curve, which is modeled using truncated regular vine copulas. By drawing simulations of the term structure from a copula-based dynamic factor model, step-ahead forecasts are obtained for zero-coupon bonds, which are then used to price both callable and non-callable fixed-coupon bonds. These bond prices are applied to solve convex multi-objective portfolio systems that account for various bond characteristics, including average returns, conditional value-at-risk, distance-to-default, transaction costs, and (option-adjusted) duration and convexity. Utilizing a sample of 879 ESG bonds from Europe over a period from January 2016 to July 2024, the MOBPs based on the proposed approach generate higher returns and Sharpe ratios compared to an equally weighted benchmark portfolio. These portfolios also result in lower tail risk, particularly during the COVID-19 pandemic and the Russo-Ukrainian war. Results further indicate that including issuer-level CO2 emissions as a portfolio attribute leads to portfolios with higher returns but also higher tail risk compared to socially responsible MOBPs based on ESG scores.
