ButterworthFilter.filter

ButterworthFilter.filter(data)

Filter a panel.

Converts the cutoff period to the penalty, then runs the shared penalized smoother column by column; a one-dimensional series is treated as a one-column panel.

Parameters:

data (ArrayLike) – A (nobs, k) panel or one-dimensional series.

Return type:

DecompositionFilterResult

Returns:

The DecompositionFilterResult, every row kept, with the first and last ceil(period) rows flagged provisional.

Raises:

• DimensionError – If the panel is malformed.

• SpecificationError – If it is shorter than 2 order + 2.

Example

>>> rng = np.random.default_rng(0)
>>> panel = np.cumsum(rng.standard_normal((150, 2)), axis=0)
>>> res = ButterworthFilter(period=32).filter(panel)
>>> res.n_series, res.method, int(res.provisional.sum())
(2, 'butterworth', 64)
>>> ButterworthFilter(order=3).filter(panel[:7, 0])
Traceback (most recent call last):
    ...
cultivars.exceptions.SpecificationError: a Butterworth filter of order 3 needs at ...