statsmodels.robust.covariance.CovM

class statsmodels.robust.covariance.CovM(data, norm_mean=None, norm_scatter=None, scale_bias=None, method='S')

M-estimator for multivariate Mean and Scatter

Interface incomplete and experimental.

Parameters:

dataarray_like

Multivariate data set with observation in rows and variables in columns.

norm_meannorm instance

If None, then TukeyBiweight norm is used. (Currently no other norms are supported for calling the initial S-estimator)

norm_scatterNone or norm instance

If norm_scatter is None, then the norm_mean will be used.

breakdown_pointfloat in (0, 0.5]

Breakdown point for first stage S-estimator.

scale_biasNone or float

Must currently be provided if norm_mean is not None.

methodstr

Currently only S-estimator has automatic selection of scale function.

Methods

fit([start_mean, start_shape, start_scale, ...])

Estimate mean, shape and scale parameters with MM-estimator

Methods

fit([start_mean, start_shape, start_scale, ...])

Estimate mean, shape and scale parameters with MM-estimator