Parametric
Stationarity test (ADF and KPSS)
Tests for a unit root (ADF) and for stationarity around a level/trend (KPSS) in a time series. The series values must be in chronological row order.
Variables
- Time seriesContinuous — 1
Options
- Regression (ADF and KPSS)No constant or trend · Constant only · Constant and trend
- Lag order (-1 = automatic)
What it reports
Checks whether a time series has a unit root or is stationary around a level/trend, using the Augmented Dickey–Fuller (ADF) and KPSS tests; the two tests’ null hypotheses are opposites.
statsmodels · Open the accuracy report
When does this apply?
For a design in which one time series is tested for a unit root (ADF) and for stationarity around a level or trend (KPSS). The series values must be in chronological row order — the software re-sorts nothing and fills no gaps. The deterministic term (the “Regression” option) and the lag order are set by you; both tests run once under that one specification, and neither configures the other. The result is diagnostic: the τ and η statistics and their p-values are reported, but no “stationary” / “not stationary” verdict is issued.