Long-Term Investing Trading Bot

Project Goals

• Learn how to use and implement a financial data API
• Create a robust program that can run autonomously with error handling
• Apply the bot to investing in order to generate a profit
• Learn about financial risk management by incorporating risk management strategies into the bot's decision-making algorithm
• Integrate machine learning and neural network algorithms for optimal trading strategies
• Gain experience with performance analysis using back testing

Our project’s objective is to develop and rigorously test a financial trading bot which can value invest in companies from a large cross section of the US market. The bot should be able to determine when to sell, should a position become overvalued or meets a stop-loss. Furthermore, the bot shall only purchase any one security up to a maximum percentage of the total portfolio. The bot will use an API to pull financial ratio data from any time period of any security and utilize this data to make an informed decision on whether or not to buy or sell a security.

Features

• Portfolio Management
  • Risk Management
  • Risk Analysis
• Machine Learning
  • Use a variety of ML models to understand market data and aid decision making process
  • Use Tensorflow neural network models to process and pick winners among the total US stock market
• Simulated Trades
  • Method of executing and storing "trades" on an accessible CSV file
  • Profit/Loss calculation on simulated trades
• Deployable
  • The bot is designed to be deployed onto the cloud for large data input and long runtimes

License

By: Vignesh Sella, Sandeep Sainath, Kevin Vasquez, and Zeeshan Kunnummal
MIT