The goal of the project is to learn the trends in stock price and perform a series of trades over a period of time and end with a profit. In each trade, Bot can either buy/sell/hold. Bot will start with an investment capital of $100,000 and it's performance is measured as a percentage of the return on investment.
Q-Learning algorithm is implemented from scratch which is used for reinforcement learning to train an agent to learn the trends in stock price and perform a series of trades.
Code and Environment exlaination, Bot performance and other information can be found in the Process.pdf
- Environment of the bot was provided by TA's of course CSE 574 at University at Buffalo in Fall 2021.