Taxable Stock Trading with Deep Reinforcement Learning

TL;DR

This paper uses deep reinforcement learning to develop stock trading strategies that incorporate tax considerations, demonstrating significant potential losses when taxes are ignored in trading decisions.

Contribution

It introduces a reinforcement learning framework that explicitly models tax effects in stock trading, highlighting the importance of tax-aware strategies.

Findings

Ignoring taxes can cause over 62% loss in portfolio returns.

Tax-aware strategies significantly outperform tax-ignorant ones.

Embedding taxes in the trading environment improves strategy effectiveness.

Abstract

In this paper, we propose stock trading based on the average tax basis. Recall that when selling stocks, capital gain should be taxed while capital loss can earn certain tax rebate. We learn the optimal trading strategies with and without considering taxes by reinforcement learning. The result shows that tax ignorance could induce more than 62% loss on the average portfolio returns, implying that taxes should be embedded in the environment of continuous stock trading on AI platforms.