Reinforcement Learning for Stocks

Comparing reinformcent learning to a manual strategy in the stock market

This was a project for a graduate CS course. Source code and written report can be shared in private upon request with potential employers.

Can a reinforcement learning algorithm like QLearner outperform a manual strategy when trading stocks in the market?

I used several technical indicators for historical stock data, a manually coded Q Learner, and a stock market simulator to backtest a learned strategy and compare it to a manual strategy.

Datasets

  • Historical stock price data.

Concepts applied

  • Q-learner, Dyna
  • Hyperparameter tuning, cross-validation, bias/variance tradeoff
  • Time-series analysis

Other Work

Sparse vs. Dense Data

Comparison of Supervised Learning Algorithms: k-NN, SVM, Decision Trees, Boosted Trees, and Neural Networks