Integration of Investor Behavioral Perspective and Climate Change in Reinforcement Learning for Portfolio Optimization

Abstract

Addressing environmental impact is increasingly imperative for individual investors and large financial institutions as one of the objectives of socially responsible investing. However, there is a noticeable gap in research on integrating sustainability and low-carbon considerations into machine learning-based portfolio optimization. To meet this challenge, this study introduces a Portfolio Emissions Aware Reinforcement Learning (PEARL) model based on the Proximal Policy Optimization (PPO) algorithm to optimize a portfolio of Dow Jones Industrial Average (DJIA) stocks. PEARL uniquely integrates environmental impact considerations, specifically carbon footprint using the firm level scope 1 and scope 2 emissions data, alongside firm-level investor sentiment and attention, into the investment decision-making process. Through multiple experiments, PEARL demonstrates significant advantages, in terms of financial and environmental performance, over the DJIA index and the standard PPO framework which focuses solely on historical prices and portfolio returns.

Keywords: socially responsible investing, investor behavior, carbon footprint, deep reinforcement learning, portfolio optimization.