statsmodels.robust.covariance.CovDetS#

class statsmodels.robust.covariance.CovDetS(data, norm=None, breakdown_point=0.5)[source]#

S-estimator for mean and covariance with deterministic starts

Parameters:

dataarray_like: Multivariate data set with observation in rows and variables in columns.
normnorm instance: If None, then TukeyBiweight norm is used. (Currently no other norms are supported for calling the initial S-estimator)
breakdown_pointfloat in (0, 0.5]: Breakdown point for first stage S-estimator.

Methods

fit(*[, h_start, mean_func, scale_func, ...])

Compute S-estimator of mean and covariance

Notes

Reproducibility: this uses deterministic starting sets and there is no randomness in the estimator. However, the estimates may not be reproducible across statsmodels versions when the methods for starting sets or default tuning parameters for the optimization change. With different starting sets, the estimate can converge to a different local optimum.

References

..[1] Hubert, Mia, Peter Rousseeuw, Dina Vanpaemel, and Tim Verdonck. 2015.: “The DetS and DetMM Estimators for Multivariate Location and Scatter.” Computational Statistics & Data Analysis 81 (January): 64-75. https://doi.org/10.1016/j.csda.2014.07.013.
..[2] Hubert, Mia, Peter J. Rousseeuw, and Tim Verdonck. 2012. “A: Deterministic Algorithm for Robust Location and Scatter.” Journal of Computational and Graphical Statistics 21 (3): 618-37. https://doi.org/10.1080/10618600.2012.672100.

Methods

fit(*[, h_start, mean_func, scale_func, ...])

Compute S-estimator of mean and covariance