lib5c.util.lowess module

Module for performing lowess fitting. Consists mostly of a convenience wrapper around statsmodels.nonparametric.smoothers_lowess.lowess().

lib5c.util.lowess.constant_fit(x, y, logx=False, logy=False, agg='median')[source]

Same signature as lowess_fit() and group_fit(), but instead of fitting y against x, simply applies an aggregating function to y.

Parameters:
  • x (Any) – Ignored, present only for signature parity with other fitters.
  • y (np.ndarray) – The y values to fit.
  • logx (Any) – Ignored, present only for signature parity with other fitters.
  • logy (bool) – Pass True to perform the fit on the scale of log(y).
  • agg ({'median', 'mean', 'lowess'}) – The function to use to aggregate y-values.
Returns:

This function takes in x values, ignores them completely, and simply returns the constant estimated y value on the original y scale (regardless of what is passed for logy).
Return type:

function
lib5c.util.lowess.group_fit(x, y, logx=False, logy=False, agg='median', left_boundary=0, right_boundary=None, n_windows=100, window_width=0.2)[source]

Simpler alternative to lowess fitting using a sliding window mean.

Parameters:
  • y (x,) – The x and y values to fit, respectively.
  • logy (logx,) – Pass True to perform the fit on the scale of log(x) and/or log(y), respectively.
  • agg ({'median', 'mean', 'lowess'}) – The function to use to aggregate within groups.
  • right_boundary (left_boundary,) – Allows specifying boundaries for the fit, in the original x space. If a float is passed, the returned fit will return the farthest left or farthest right lowess-estimated y_hat (from the original fitting set) for all points which are left or right of the specified left or right boundary point, respectively.
  • n_windows (int) – The number of windows to use (spaced uniformly across the range of x).
  • window_width (float) – The width of each window, defined as a fraction of its x-value.
Returns:

This function takes in x values on the original x scale and returns estimated y values on the original y scale (regardless of what is passed for logx and logy). This function will still return sane estimates for y even at points not in the original fitting set by performing linear interpolation in the space the fit was performed in.
Return type:

function
lib5c.util.lowess.lowess_agg(y, it=3)[source]

Performs an aggregation operation equivalent to lowess. Should behave like an outlier-resistant mean.

Parameters:
  • y (np.ndarray) – The values to aggregate.
  • it (int) – The number of residual-based reweightings to perform.
Returns:

The lowess-implemented outlier-resistant mean.
Return type:

float
lib5c.util.lowess.lowess_fit(x, y, logx=False, logy=False, left_boundary=0, right_boundary=None, frac=0.3, delta=0.01)[source]

Opinionated convenience wrapper for lowess smoothing.

Parameters:
  • y (x,) – The x and y values to fit, respectively.
  • logy (logx,) – Pass True to perform the fit on the scale of log(x) and/or log(y), respectively.
  • right_boundary (left_boundary,) – Allows specifying boundaries for the fit, in the original x space. If a float is passed, the returned fit will return the farthest left or farthest right lowess-estimated y_hat (from the original fitting set) for all points which are left or right of the specified left or right boundary point, respectively.
  • frac (float) – The lowess smoothing fraction to use.
  • delta (float) – Distance (on the scale of x or log(x)) within which to use linear interpolation when constructing the initial fit, expressed as a fraction of the range of x or log(x).
Returns:

This function takes in x values on the original x scale and returns estimated y values on the original y scale (regardless of what is passed for logx and logy). This function will still return sane estimates for y even at points not in the original fitting set by performing linear interpolation in the space the fit was performed in.
Return type:

function

Notes

No filtering of input values is performed; clients are expected to handle this if desired. NaN values should not break the function, but x points with zero values passed when logx is True are expected to break the function.

Linear interpolation between x-values in the original fitting set is used to provide a familiar functional interface to the fitted function.

The default value of the delta parameter is set to be non-zero, matching the behavior of lowess smoothing in R and improving performance.

Boundary conditions on the fitted function are exposed, mostly as a convenience for points where x == 0 when fitting was performed on the scale of log(x).