Automated stock trading using Deep Reinforcement Learning with Fundamental Indicators

A summary of my final project in the alumni-mentored research project at Columbia University in Summer 2021: Application of Reinforcement Learning to Finance

Photo by Austin Distel on Unsplash

April 4th, 2022 Update:
The codes in the post and my notebook might not work anymore due to FinRL’s continuous updates. So please make sure to check out the latest information on FinRL’s GitHub when you test the idea of this post. Thank you so much for your interest!

This blog post explains how I trained a deep reinforcement learning model with companies’ fundamental indicators and how well-performed the trained model was compared to a benchmark. I used the FinRL library developed by AI4Finance to train the model. This project is inspired and based on the tutorial of FinRL authored by Bruce Yang: FinRL for Quantitative Finance: Tutorial for Multiple Stock Trading.

My google Colaboratory notebook for this project is available on my GitHub.