PROTECTED SOURCE SCRIPT
QUANT - LAB ADF-GLS + COINT + VRT-WB [ERS]
ADF-GLS + COINT + VRT-WB [ERS] V9.2 INSTITUTIONAL
Institutional-grade econometric suite for unit root testing, cointegration analysis, and mean-reversion detection.
Unit Root Tests:
ADF-GLS (Elliott, Rothenberg & Stock, 1996) with MAIC lag selection
Phillips-Perron Z_t with Newey-West correction
KPSS stationarity test (confirmatory)
MZ-alpha test
Cointegration (Bivariate):
Engle-Granger two-step test (MacKinnon 2010 critical values)
Johansen Trace test (Osterwald-Lenum 1992 CVs)
Real-time spread Z-score with tick-by-tick updates
Mean-Reversion:
Variance Ratio Test (Lo-MacKinlay 1988)
Mammen Wild Bootstrap for heteroskedasticity robustness
Half-life estimation with 95% CI (delta method)
Diagnostics:
Ljung-Box Q(4) for residual autocorrelation
ARCH(4) test for heteroskedasticity
HAC standard errors (Newey-West)
Important:
Screening tool only — validate in Python/R/statsmodels
Beta SE is BIASED (generated regressor problem)
Johansen limited to bivariate systems
Bootstrap p-value resolution ~2-5%
NOT a trading system
References: ERS (1996), Lo & MacKinlay (1988), Engle & Granger (1987), Johansen (1988), MacKinnon (2010)
Institutional-grade econometric suite for unit root testing, cointegration analysis, and mean-reversion detection.
Unit Root Tests:
ADF-GLS (Elliott, Rothenberg & Stock, 1996) with MAIC lag selection
Phillips-Perron Z_t with Newey-West correction
KPSS stationarity test (confirmatory)
MZ-alpha test
Cointegration (Bivariate):
Engle-Granger two-step test (MacKinnon 2010 critical values)
Johansen Trace test (Osterwald-Lenum 1992 CVs)
Real-time spread Z-score with tick-by-tick updates
Mean-Reversion:
Variance Ratio Test (Lo-MacKinlay 1988)
Mammen Wild Bootstrap for heteroskedasticity robustness
Half-life estimation with 95% CI (delta method)
Diagnostics:
Ljung-Box Q(4) for residual autocorrelation
ARCH(4) test for heteroskedasticity
HAC standard errors (Newey-West)
Important:
Screening tool only — validate in Python/R/statsmodels
Beta SE is BIASED (generated regressor problem)
Johansen limited to bivariate systems
Bootstrap p-value resolution ~2-5%
NOT a trading system
References: ERS (1996), Lo & MacKinlay (1988), Engle & Granger (1987), Johansen (1988), MacKinnon (2010)
Protected script
This script is published as closed-source. However, you can use it freely and without any limitations – learn more here.
Institutional-grade diagnostics: GARCH, HMM Regimes, Cointegration, Microstructure, Fractal Analysis | Research only
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.
Protected script
This script is published as closed-source. However, you can use it freely and without any limitations – learn more here.
Institutional-grade diagnostics: GARCH, HMM Regimes, Cointegration, Microstructure, Fractal Analysis | Research only
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.