pmdarima.arima.KPSSTest

class pmdarima.arima.KPSSTest(alpha=0.05, null='level', lshort=True)[source][source]

Conduct a KPSS test for stationarity.

In econometrics, Kwiatkowski–Phillips–Schmidt–Shin (KPSS) tests are used for testing a null hypothesis that an observable time series is stationary around a deterministic trend (i.e. trend-stationary) against the alternative of a unit root.

Parameters:

alpha : float, optional (default=0.05)

Level of the test

null : str, optional (default=’level’)

Whether to fit the linear model on the one vector, or an arange. If null is ‘trend’, a linear model is fit on an arange, if ‘level’, it is fit on the one vector.

lshort : bool, optional (default=True)

Whether or not to truncate the l value in the C code.

Notes

This test is generally used indirectly via the pmdarima.arima.ndiffs() function, which computes the differencing term, d.

References

[R61]R’s tseries KPSS test source code: http://bit.ly/2eJP1IU

Methods

get_params([deep]) Get parameters for this estimator.
is_stationary(x) Test whether the time series is stationary.
set_params(**params) Set the parameters of this estimator.
should_diff(x) Test whether the time series is stationary or needs differencing.
__init__(alpha=0.05, null='level', lshort=True)[source][source]

Initialize self. See help(type(self)) for accurate signature.

get_params(deep=True)[source]

Get parameters for this estimator.

Parameters:

deep : bool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

params : mapping of string to any

Parameter names mapped to their values.

is_stationary(x)[source]

Test whether the time series is stationary.

Parameters:

x : array-like, shape=(n_samples,)

The time series vector.

Returns:

pval : float

The computed P-value of the test.

sig : bool

Whether the P-value is significant at the alpha level. More directly, whether to difference the time series.

set_params(**params)[source]

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**params : dict

Estimator parameters.

Returns:

self : object

Estimator instance.

should_diff(x)[source][source]

Test whether the time series is stationary or needs differencing.

Parameters:

x : array-like, shape=(n_samples,)

The time series vector.

Returns:

pval : float

The computed P-value of the test.

sig : bool

Whether the P-value is significant at the alpha level. More directly, whether to difference the time series.