This project performs large scale Monte Carlo simulations of the eigenvalues that appear in Johansen's null distribution. Results are written to `*.dat` files which can be reloaded later for analysis.
Build a forecasting model to predict the sale of a store.
Project exploring the possibility of building a statistical arbitrage strategy based on pricing inefficiencies between three S&P 500 ETFs. The approach combines Johansen cointegration analysis to define mean-reverting spreads, and ARMA forecasting to generate trading signals.
Statistical arbitrage engine that screens S&P 500 pairs using Engle-Granger and Johansen cointegration tests, fits Ornstein-Uhlenbeck dynamics via MLE, and trades spreads with a Kalman filter hedge ratio. Includes a walk-forward backtest with monthly pair re-screening, continuous position carry-over, and a full performance dashboard.
An advanced mean-reversion trading strategy for ETF baskets using Bayesian Optimization to maximize Sharpe Ratio. Features walk-forward analysis, cointegration testing, and comprehensive backtesting reports.
Code from the paper Over the Stable Space of a Multivariate Time Series
Walk-forward study of equity cointegration persistence using matched controls, multiple-testing correction, frozen-relationship validation, and predeclared robustness analysis.
VECM analysis of the 3-2-1 crack spread on 1,361 weekly WTI, NY Harbor gasoline and heating oil observations (2000-2026): Johansen cointegration, error-correction loadings and Bonferroni-corrected Granger causality test whether crude cost-pushes refined products or gasoline demand pulls crude.
ARDL cointegration in Python: three-test bounds testing with degeneracy classification, response-surface critical values, bootstrap inference, NARDL, QARDL, Fourier-ADL, and heterogeneous panels (MG/PMG/CS-ARDL) — validated against R and Stata.
To associate your repository with the johansen-test topic, visit your repo's landing page and select "manage topics."