incline.SavitzkyGolay

class incline.SavitzkyGolay(window_length=15, degree=3)

Savitzky-Golay filter: local polynomial least squares on a fixed window.

A fixed convolution, so the derivative operator is available in closed form as well as by probing – the two agree to 1e-15, which makes this the natural cross-check on the probe machinery.

Variables:

• window_length (int) – Odd number of points in the window.

• degree (int) – Degree of the local polynomial.

Parameters:

• window_length (int)

• degree (int)

__init__(window_length=15, degree=3)

Parameters:

• window_length (int)

• degree (int)

Return type:

None

Methods

__init__([window_length, degree])
analytic_operators(axis, derivative_order)	State the smoothing and derivative operators directly, if known.
bootstrap_uncertainty(estimate, axis, y, ...)	Bootstrap a nonlinear smoother, preserving short-range dependence.
closed_form_se(axis, derivative_order, ...)	Interior standard error from the filter coefficients directly.
evaluate(axis, y, derivative_order)	Filter the series and differentiate.
evaluate_with_noise(axis, y, ...)	Evaluate, allowing an adaptive smoother to use a fitted covariance.
fit(axis, y[, derivative_order, ...])	Estimate the trend and, optionally, its uncertainty.
native_posterior(axis, y, derivative_order, ...)	Uncertainty from the smoother's own probability model.
operators(axis, derivative_order)	The smoothing and derivative operators for this configuration.
params()	Report the window and polynomial degree.
scale_of(axis)	Window length as a fraction of the sample.
with_scale(scale, axis)	Set the window to scale of the sample.

Attributes

degree
has_native_posterior
is_linear	Whether the derivative is a fixed linear map of the data.
linear
name
requires_regular_grid
supported_orders
uses_noise_for_fit	Whether a supplied noise model can change the point estimate.
window_length

name: ClassVar[str] = 'sgolay'

linear: ClassVar[bool] = True

supported_orders: ClassVar[frozenset[int]] = frozenset({0, 1, 2, 3})

requires_regular_grid: ClassVar[bool] = True

window_length: int = 15

degree: int = 3

evaluate(axis, y, derivative_order)

Filter the series and differentiate.

Parameters:

• axis (TimeAxis)

• y (npt.NDArray[np.float64])

• derivative_order (int)

Return type:

Evaluation

closed_form_se(axis, derivative_order, standard_deviation)

Interior standard error from the filter coefficients directly.

Parameters:

• axis (TimeAxis) – The time axis.

• derivative_order (int) – Derivative derivative_order.

• standard_deviation (float) – Noise standard deviation.

Returns:

The constant interior standard error, broadcast over the series.

Return type:

npt.NDArray[np.float64]

with_scale(scale, axis)

Set the window to scale of the sample.

Parameters:

• scale (float)

• axis (TimeAxis)

Return type:

Self

scale_of(axis)

Window length as a fraction of the sample.

Parameters:

axis (TimeAxis)

Return type:

float

params()

Report the window and polynomial degree.

Return type:

dict[str, Any]