"Engineering Alpha: Filtering Trading Signals from Market Noise"
In increasingly competitive markets, the edge lies not in more data, but in better data and sharper filters. Engineering Alpha: Filtering Trading Signals from Market Noise is written for quantitative researchers, systematic portfolio managers, advanced practitioners, and technically minded traders who want to turn fragile backtests into durable trading strategies. It bridges the gap between academic theory and production-grade implementation, showing how to transform noisy price streams into robust signals that can survive the real world of liquidity constraints, costs, and regime shifts.
The book develops a complete research stack, from mathematical foundations and time-series modeling through feature engineering, machine learning, and factor construction. Readers will learn to clean and align market and alternative data, design leak-free targets, and apply filters such as ARIMA, Kalman, and wavelet-based methods. It then formalizes validation via walk-forward testing, purged cross-validation, multiple-testing control, and performance metrics like Sharpe, Information Ratio, and IC-decay. Finally, it connects signals to portfolios-covering risk models, constrained optimization, position sizing, execution algorithms, and monitoring-so that estimated alpha translates into risk-adjusted PnL.
The material assumes comfort with Python, basic linear algebra, and probability, but it is self-contained where it matters for practice. Throughout, the emphasis is on reproducible workflows, time-aware evaluation, and eng
