Skip to content
← Projects

Quant · Dec 2025

Statistical Pairs Trading Backtest

An Engle-Granger market-neutral strategy applied to AAPL/MSFT, KO/PEP, and XOM/CVX over daily data from 2015 to 2023. The backtest incorporates walk-forward cointegration screening, out-of-sample hedge ratios, and a 1 bp transaction cost assumption. Sharpe ratio, maximum drawdown, turnover, and signal decay are reported.

AAPL/MSFT, KO/PEP, XOM/CVX

Pairs

2015 to 2023 daily

Window

1 bp

Costs

Walk-forward

Method

Problem

Many publicly available pairs trading studies introduce look-ahead bias through full-sample hedge ratios. I aimed to produce an honest evaluation that accounts for realistic transaction costs.

Approach

I applied Engle-Granger cointegration screening using rolling training and testing windows, computed out-of-sample hedge ratios at each refit, and assumed a 1 bp transaction cost on every fill. I tracked Sharpe ratio, maximum drawdown, cumulative return, turnover, and signal decay throughout.

Results

The analysis produced a clear out-of-sample picture of which pair relationships persist, where transaction costs erode the edge, and how quickly signals decay. Results are reported transparently, including cases in which the strategy underperforms.

Stack

PythonstatsmodelsyfinancePandas