Unit Root Testing¶

Many time series are highly persistent, and determining whether the data appear to be stationary or contains a unit root is the first step in many analyses. This module contains a number of routines:

• Augmented Dickey-Fuller (ADF)

• Dickey-Fuller GLS (DFGLS)

• Phillips-Perron (PhillipsPerron)

• KPSS (KPSS)

• Zivot-Andrews (ZivotAndrews)

• Variance Ratio (VarianceRatio)

• Automatic Bandwidth Selection (auto_bandwidth())

The first four all start with the null of a unit root and have an alternative of a stationary process. The final test, KPSS, has a null of a stationary process with an alternative of a unit root.

Cointegration Analysis¶

The module extended the single-series unit root testing to multiple series and cointegration testing and cointegrating vector estimation.

• Cointegrating Testing

  • Engle-Granger Test (engle_granger)

  • Phillips-Ouliaris Tests (phillips_ouliaris)

• Cointegrating Vector Estimation

  • Dynamic OLS (DynamicOLS)

  • Fully Modified OLS (FullyModifiedOLS)

  • Canonical Cointegrating Regression (CanonicalCointegratingReg)