Robust Linear Models Link to Notebook GitHub In [1]: from __future__ import print_function import numpy as np import statsmodels.api as sm import matplotlib.pyplot as plt from statsmodels.sandbox.regression.predstd import wls_prediction_std Estimation Load data: In [2]: data = sm.datasets.stackloss.load() data.exog = sm.add_constant(data.exog) Huber's T norm with the (default) median absolute deviation scaling In [3]: huber_t = sm.RLM(data.endog, data.e

statsmodels.sandbox.distributions.extras.NormExpan_gen.entropy NormExpan_gen.entropy(*args, **kwds) Differential entropy of the RV. Parameters: arg1, arg2, arg3,... : array_like The shape parameter(s) for the distribution (see docstring of the instance object for more information). loc : array_like, optional Location parameter (default=0). scale : array_like, optional Scale parameter (default=1).

statsmodels.tsa.vector_ar.irf.IRAnalysis.lr_effect_stderr IRAnalysis.lr_effect_stderr(orth=False) [source]

statsmodels.discrete.discrete_model.NegativeBinomialResults.normalized_cov_params NegativeBinomialResults.normalized_cov_params()

statsmodels.tsa.vector_ar.var_model.VARProcess.plot_acorr VARProcess.plot_acorr(nlags=10, linewidth=8) [source] Plot theoretical autocorrelation function

statsmodels.sandbox.tsa.fftarma.ArmaFft.spd ArmaFft.spd(npos) [source] raw spectral density, returns Fourier transform n is number of points in positive spectrum, the actual number of points is twice as large. different from other spd methods with fft

statsmodels.tsa.vector_ar.dynamic.DynamicVAR.resid static DynamicVAR.resid() [source]

statsmodels.sandbox.distributions.extras.NormExpan_gen.est_loc_scale NormExpan_gen.est_loc_scale(*args, **kwds) est_loc_scale is deprecated! This function is deprecated, use self.fit_loc_scale(data) instead.

statsmodels.regression.linear_model.RegressionResults.bic static RegressionResults.bic() [source]

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.fit NonlinearIVGMM.fit(start_params=None, maxiter=10, inv_weights=None, weights_method='cov', wargs=(), has_optimal_weights=True, optim_method='bfgs', optim_args=None) Estimate parameters using GMM and return GMMResults TODO: weight and covariance arguments still need to be made consistent with similar options in other models, see RegressionResult.get_robustcov_results Parameters: start_params : array (optional) starting value for parameter