This project aims to implement Reinforcement Learning algorithms for financial data analysis and forecasting. Time series stock data is fetched from Yahoo Finance API and analyzed using the Upper Confidence Bound (UCB) algorithm. The confidence interval is displayed in the Angular frontend. Forecasting functionality is in development, using Vector Auto Regression (VAR) and Keras Long Short-Term Memory (LSTM) models.
- Upper Confidence Bound (UCB): Reinforcement Learning algorithm to analyze financial data and create recommendations based on risk tolerance and historical performance.
- Angular Frontend: Interactive user interface to search for stocks and display the confidence bounds.
- VAR Forecasting: Utilizes the Vector Auto Regression model to forecast financial trends.
- Keras LSTM Forecasting: Incorporates the Long Short-Term Memory model from Keras for time series forecasting.
- Clone the repository.
- Install the required packages using
pip install -r requirements.txt. - Run the backend using
python server.py.
- cd into "UCB-frontend".
- Install the required packages using
npm install. - Run the frontend using
ng serve. - The frontend will be running on
http://localhost:4200.
A demo of the application is yet to be deployed.
