Back to models
Param Estimator

StationarityKPSS

API referenceSource

Test for stationarity via the Kwiatkowski-Phillips-Schmidt-Shin Test.

Quickstart

python
from sktime.param_est.stationarity import StationarityKPSS

estimator = StationarityKPSS(p_threshold=0.05, regression='c', nlags='auto')

Parameters(3)

p_thresholdfloat, optional, default=0.05
significance threshold to apply in testing for stationarity
regressionstr, one of {“c”,”ct”,”ctt”,”n”}, optional, default=”c”

Constant and trend order to include in regression.

  • “c”: constant only (default).

  • “ct”: constant and trend.

nlagsstr or int, optional, default=”auto”. If int, must be positive.

Indicates the number of lags to be used internally in kpss. If “auto”, lags is calculated using the data-dependent method of Hobijn et al (1998). See also Andrews (1991), Newey & West (1994), and Schwert (1989). If “legacy”, uses int(12 * (n / 100)**(1 / 4)), as outlined in Schwert (1989). If int, uses that exact number.

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.param_est.stationarity import StationarityKPSS
>>> 
>>> X = load_airline ()
>>> sty_est = StationarityKPSS ()
>>> sty_est. fit (X) StationarityKPSS(
... )
>>> sty_est. get_fitted_params ()["stationary" ] False