Transformer

ConditionalDeseasonalizer

Remove seasonal components from time series, conditional on seasonality test.

Fit tests for seasonality and if the passed time series has a seasonal component it applies seasonal decomposition provided by statsmodels <https://www.statsmodels.org> to compute the seasonal component. If the test is negative _seasonal is set to all ones (if model is “multiplicative”) or to all zeros (if model is “additive”).

Transform aligns seasonal components stored in seasonal_ with the time index of the passed series and then subtracts them (“additive” model) from the passed series or divides the passed series by them (“multiplicative” model).

Quickstart

python

from sktime.transformations.detrend import ConditionalDeseasonalizer

estimator = ConditionalDeseasonalizer(seasonality_test=None, sp=1, model='additive')

Parameters(3)

seasonality_testcallable or None, default=None: Callable that tests for seasonality and returns True when data is seasonal and False otherwise. If None, 90% autocorrelation seasonality test is used.
spint, default=1: Seasonal periodicity.
model{“additive”, “multiplicative”}, default=”additive”: Model to use for estimating seasonal component.

Examples

>>> from sktime.transformations.detrend import ConditionalDeseasonalizer
>>> from sktime.datasets import load_airline
>>> y = load_airline ()
>>> transformer = ConditionalDeseasonalizer (sp = 12)
>>> y_hat = transformer. fit_transform (y)