01 — PROBLEM STATEMENT
Traditional investment strategies often fail to account for dynamic market conditions and correlation shifts. Manual portfolio rebalancing is inefficient and prone to emotional bias. There was a need for data-driven, systematic approaches to portfolio construction and risk management.
02 — METHODOLOGY
Implemented Modern Portfolio Theory (MPT) to construct efficient frontiers using historical return data from Yahoo Finance API.
Built a Monte Carlo simulation engine to model 10,000+ portfolio weight combinations and identify optimal risk-return profiles.
Developed ARIMA and GARCH models for time-series forecasting of asset prices and volatility.
Created a risk management dashboard using Python (Pandas, Matplotlib) to visualize Value at Risk (VaR) and Conditional VaR.
Backtested strategies over 5-year periods, accounting for transaction costs and slippage.
Automated portfolio rebalancing logic based on threshold triggers (e.g., weight drift > 5%).
Integrated real-time data feeds to monitor portfolio performance and generate alerts.
03 — OUTCOMES & IMPACT
Achieved a Sharpe Ratio of 1.85 on a diversified equity portfolio, outperforming the S&P 500 benchmark by 12% over the test period.
Reduced portfolio volatility by 18% through systematic diversification and correlation analysis.
Time-series models achieved 92% accuracy in short-term price movement predictions.
Built a reusable Python framework for portfolio analysis, now used by peers in finance coursework.
Presented findings in a university seminar on "Quantitative Methods in Asset Management."
Developed strong proficiency in financial engineering and statistical modeling.