In this project, I collaborated with a team to backtest financial models using historical stock data from 500 companies (1985–2021) to evaluate the predictive power of key investment factors: Book-to-Price Ratio, Return Volatility, and 3-Month Price Momentum.

Process & Outcomes

• Data Preparation: Cleaned and organized the dataset, creating deciles for each factor to measure their performance against median benchmark returns.

• Factor Analysis: Visualized factor distributions, assessed predictive performance, and tested inter-factor correlations to identify redundancies.

• Model Creation: Developed a multi-factor stock selection model by combining Return Volatility (55%) and Book-to-Price (45%) for optimal performance.

• Results:

  • The final model achieved an 18.25% long-short return and consistently outperformed individual factors.

  • Demonstrated strong predictive accuracy and insights for portfolio optimization.

This project refined my quantitative analysis skills and ability to develop data-driven financial models while leading the team to deliver actionable investment insights through organized workflows and clear execution.