Skip to content

Cointegration is the key trigger in statistical arbitrage for correlated pair, measuring inconsistency and stability properties

Notifications You must be signed in to change notification settings

ventositwaitang/A-Cointegration-Learning

Repository files navigation

Cointegration

Cointegration is the key trigger in statistical arbitrage for correlated pair, measuring inconsistency and stability properties

Navigation

Explanation of Homoscedasticity with normal diffused (white noise)

ADF_Cointegration.ipynb

diffusion convergence with drift term

KPSS_Cointegration.ipynb

Thesis: A Pairs Causality Research with ARIMA Forecasting

Future direction

  • to be refined by Grid Search for time-window & stop-profit hyper-param, and calibrating divergent adjusting factor (Cocoercivity, a basin hopping parameter for long-term drift)

  • adding momentum rules(MA cross/ RSI/ tailormade Alpha) in practice

  • as Multi-asset Pricing Model by Second-Order Cone Programming (Solver for Linear/ Quadratic(non-linear) Gaussian simulation) with neural autograd

About

Cointegration is the key trigger in statistical arbitrage for correlated pair, measuring inconsistency and stability properties

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published