statsmodels.miscmodels.count.PoissonZiGMLE.score PoissonZiGMLE.score(params) Gradient of log-likelihood evaluated at params

statsmodels.regression.linear_model.OLSResults.bic static OLSResults.bic()

statsmodels.sandbox.regression.gmm.IVRegressionResults.HC3_se static IVRegressionResults.HC3_se() See statsmodels.RegressionResults

statsmodels.regression.linear_model.OLSResults.scale static OLSResults.scale()

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

statsmodels.sandbox.regression.gmm.IVGMMResults.ssr static IVGMMResults.ssr() [source]

statsmodels.regression.linear_model.OLSResults.cov_HC1 static OLSResults.cov_HC1() See statsmodels.RegressionResults

statsmodels.regression.linear_model.GLS.information GLS.information(params) Fisher information matrix of model Returns -Hessian of loglike evaluated at params.

statsmodels.sandbox.distributions.transformed.ExpTransf_gen class statsmodels.sandbox.distributions.transformed.ExpTransf_gen(kls, *args, **kwargs) [source] Distribution based on log/exp transformation the constructor can be called with a distribution class and generates the distribution of the transformed random variable Methods cdf(x, *args, **kwds) Cumulative distribution function of the given RV. entropy(*args, **kwds) Differential entropy of the RV. est_loc_scale(*args, **kwds) est_

Vector Autoregressions tsa.vector_ar VAR(p) processes We are interested in modeling a multivariate time series , where denotes the number of observations and the number of variables. One way of estimating relationships between the time series and their lagged values is the vector autoregression process: where is a coefficient matrix. We follow in large part the methods and notation of Lutkepohl (2005), which we will not develop here. Model fitting Note The classes referenced below ar