statsmodels.robust.covariance.CovDetS.fit#

CovDetS.fit(*, h_start=None, mean_func=None, scale_func=None, maxiter=100, options_start=None, maxiter_step=5)[source]#

Compute S-estimator of mean and covariance

Parameters:

h_startint: Number of observations used in starting mean and covariance.
mean_func, scale_funccallable or None: Mean and scale function for initial standardization. Current defaults, if they are None, are median and mad, but default scale_func will likely change.
maxiterint: Maximum number of iterations for the c-step of the best candidate solution.
options_startNone or dict: Options for the starting estimators. TODO: which options? e.g., for OGK
maxiter_stepint: Number of iterations used for each starting candidate before selecting the best one for further iteration.

Returns:

CovMResult: Named tuple with mean, shape, scale, cov and extra attributes scale_all, idx_best, tmean, tscale from the starting-set search. See CovMResult for details.