statsmodels.nonparametric.bandwidths.bw_scott statsmodels.nonparametric.bandwidths.bw_scott(x, kernel=None) [source] Scott?s Rule of Thumb Parameters: x : array-like Array for which to get the bandwidth kernel : CustomKernel object Unused Returns: bw : float The estimate of the bandwidth Notes Returns 1.059 * A * n ** (-1/5.) where A = min(std(x, ddof=1), IQR/1.349) IQR = np.subtract.reduce(np.percentile(x, [75,25])) References Scott, D.W. (1992) Multivariate Density Estimation: T

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.score_cu NonlinearIVGMM.score_cu(params, epsilon=None, centered=True)

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.start_weights NonlinearIVGMM.start_weights(inv=True)

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.predict NonlinearIVGMM.predict(params, exog=None) [source]

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.score NonlinearIVGMM.score(params, weights, **kwds) [source]

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.momcond NonlinearIVGMM.momcond(params)

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.momcond_mean NonlinearIVGMM.momcond_mean(params) mean of moment conditions,

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.jac_func NonlinearIVGMM.jac_func(params, weights, args=None, centered=True, epsilon=None) [source]

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.gmmobjective_cu NonlinearIVGMM.gmmobjective_cu(params, weights_method='cov', wargs=()) objective function for continuously updating GMM minimization Parameters: params : array parameter values at which objective is evaluated Returns: jval : float value of objective function

statsmodels.sandbox.regression.gmm.NonlinearIVGMM.gradient_momcond NonlinearIVGMM.gradient_momcond(params, epsilon=0.0001, centered=True) gradient of moment conditions Parameters: params : ndarray parameter at which the moment conditions are evaluated epsilon : float stepsize for finite difference calculation centered : bool This refers to the finite difference calculation. If centered is true, then the centered finite difference calculation is used. Otherwise the one-sided forward dif