* Implemented VWAP and TWAP execution algorithms using Python and Machine Learning models for predictive modeling and statistical analysis.* Integrated high-frequency trading data (price, volume, time intervals) for 50 stocks over a 6-month period, using Pandas for efficient data manipulation and scikit-learn for time series forecasting to optimize trade execution.* Employed regression analysis and time series models (ARIMA, GARCH) to forecast intra-day volume patterns, improving algorithmic precision and adapting to market conditions in real-time.* Key Achievements:* Reduced market impact by 9% : Implemented the project’s findings into the firm’s live trading system, improving overall trading performance.* Reduced slippage by 6% across algorithmic trading strategies: Enhanced execution efficiency by implementing advanced VWAP and TWAP algorithms, outperforming manual order execution.* Lowered execution costs by 5% within 3 months: Optimized the volume distribution curve based on historical trade volume patterns, leading to a significant reduction in overall execution expenses.

* Implemented factor-based quantitative trading models using Python, employing Random Forest, Gradient Boosting while fine-tuning the parameters.* Conducted analysis of microstructure data, including price, volume, and volatility, for over 100 stocks over a 12-month period, identifying significant return-driving factors such as momentum, size, value, and liquidity.* Utilized feature selection methods (LASSO, Ridge regression) to filter out noise and retain the relevant factors (earnings surprises, short-term reversal).* Discovered new alpha-generating factors by analyzing cross-sectional stock data through XGBoost and random forest classifiers, achieving a 12% increase in annualized returns on large-cap stocks.* Key Achievements:* Discovered 5 new predictive factors: Enhanced strategy performance by incorporating machine learning-discovered factors, price momentum decay and volatility clustering, into multi-factor models, boosting alpha by 15% annually.* Improved factor robustness by 20%: Implemented rolling window backtesting and cross-validation to ensure stability and out-of-sample performance across market regimes.*Increased portfolio risk-adjusted returns by 9%: Optimized factor loadings through mean-variance optimization.