incline.trend_with_sizer

incline.trend_with_sizer(df, value_column='value', time_column=None, smoother=None, min_persistence=3, *, scales=None, n_scales=20, scale_range=(0.02, 0.5), confidence_level=0.95, noise=None, simultaneous=True, n_bootstrap=200, random_state=None)[source]

Estimate a trend and mark where it survives a scale sweep.

Parameters:

  • df (pd.DataFrame) – Time series data.

  • value_column (str) – Column holding the values.

  • time_column (str | None) – Numeric time column.

  • smoother (Smoother | None) – Estimator used for both the trend and the sweep.

  • min_persistence (int) – Consecutive scales required to call a feature real.

  • scales (npt.NDArray[np.float64] | None) – Explicit smoothing scales, or None to generate them.

  • n_scales (int) – Number of generated scales.

  • scale_range (tuple[float, float]) – Lower and upper generated scales.

  • confidence_level (float) – Confidence level for both the trend and the map.

  • noise (NoiseModel | str | None) – Noise model used by both the trend and the map.

  • simultaneous (bool) – Whether both requests use whole-curve bands.

  • n_bootstrap (int) – Bootstrap replicates for nonlinear smoothers.

  • random_state (int | np.random.Generator | None) – Seed or Generator controlling all bootstrap draws.

Returns:

The estimator’s usual columns plus sizer_significance, persistent_increasing and persistent_decreasing.

Return type:

pd.DataFrame