Inspiration

Predicting stock market performance is a centuries old problem. Ambitious investors have designed and redesigned thousands of trading algorithms, custom indexes, and more to gain an edge in market prediction.

What it does

Use reinforcement learning algorithms to model trading stocks like a game and train an agent to maximize the reward function we care about: total money gained.

How I built it

Challenges I ran into

Accomplishments that I'm proud of

Trained an agent to make a trading decision (buy, sell, hold) to execute at the opening of the following day.

Built With

Share this project:

Updates