Pairs Trading with Kalman Filters: Dynamic Hedge Ratios
Why a static OLS beta falls short: estimating a time-varying hedge ratio with a state-space model and building a cointegration strategy.
archive
All Posts
Notes on algorithmic strategy research, backtesting discipline, data quality and quantitative thinking.
Do Transformers Actually Work on Financial Time Series?
Benchmarking attention against LSTM and classical ARIMA for return forecasting. The results may surprise you.
Volatility Forecasting with GARCH: A BIST 100 Case Study
From testing ARCH effects to interpreting GARCH(1,1) parameters: modelling volatility clustering step by step.
Stationarity Tests: ADF, KPSS and Why You Should Use Both
The logic of unit-root tests, their power problems, and the most common mistakes with price series.
The Silent Killers of Backtests: Look-Ahead and Survivorship Bias
Why do strategies that look great on paper collapse live? Common methodological errors, shown with example code.
Cleaning Financial Data: Missing Bars, Splits and Dividend Adjustments
Getting raw price data model-ready: an end-to-end cleaning pipeline with pandas.
Vectorization with NumPy: Make Your Backtest 100x Faster
From for-loops to broadcasting: writing the same strategy three ways and timing each.
Paper Digest — Momentum 30 Years On: Rereading Jegadeesh & Titman
Do the classic momentum findings still hold on current data? A replication and discussion.
Feature Engineering: Turning Financial Time Series into Model Inputs
Return-, volatility- and microstructure-based features — and how to build them without leakage.
Intro to Portfolio Optimization: From Markowitz to Risk-Based Weighting
The theory of mean-variance optimization, its fragilities and practical alternatives.