Design-Based Research Capstone

Capstone Brief - Quantitative Trading Bot

A CSA capstone project using machine learning, news sentiment analysis, database systems, and graph-based data structures to predict short-term stock market movement through an iterative DBR development process. The system is also being integrated into Fortune Finders and Wand to create an interactive market prediction game.

View Project

In Progress

Project Leads Anvay, Sai, Aashray

Quantitative Trading Bot

ML + News Sentiment for short-term market prediction (DBR Capstone)

✓ Combines price data with real-time news sentiment

✓ Built using iterative Design-Based Research (DBR) cycles

✓ Uses ML + feature engineering for prediction

✓ Backtesting + paper trading planned for evaluation

✓ Integrated with Fortune Finders and Wand to create an interactive trading game

Python Machine Learning NLP Sentiment Web Scraping SQL Database Graphs Backtesting Alpaca API (planned) Interactive Game Integration

About

We are developing a quantitative trading bot that predicts short-term stock movement using market indicators and real-time financial news sentiment. The project follows a DBR approach with repeated build-test-refine cycles to improve performance. The bot will also power an interactive trading experience through integration with Fortune Finders and Wand, allowing users to test predictions and strategies in a game-like environment.

Impact

Makes professional-style quant workflows accessible to students

Creates a structured dataset of price/news/sentiment/predictions

Supports model evaluation through backtesting

Enables an interactive trading simulation through Fortune Finders and Wand

Designed for deployment through OCS capstone pages

Learn More