statsmodels.tsa.vector_ar.var_model.VARResults.cov_ybar VARResults.cov_ybar() [source] Asymptotically consistent estimate of covariance of the sample mean Notes Lutkepohl Proposition 3.3

statsmodels.tsa.vector_ar.var_model.VARResults.fevd VARResults.fevd(periods=10, var_decomp=None) [source] Compute forecast error variance decomposition (?fevd?) Returns: fevd : FEVD instance

statsmodels.tsa.vector_ar.var_model.VARResults.acf VARResults.acf(nlags=None) Compute theoretical autocovariance function Returns: acf : ndarray (p x k x k)

statsmodels.tsa.vector_ar.var_model.VARResults.acorr VARResults.acorr(nlags=None) Compute theoretical autocorrelation function Returns: acorr : ndarray (p x k x k)

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

statsmodels.tsa.vector_ar.var_model.VARProcess.orth_ma_rep VARProcess.orth_ma_rep(maxn=10, P=None) [source] Compute Orthogonalized MA coefficient matrices using P matrix such that . P defaults to the Cholesky decomposition of Parameters: maxn : int Number of coefficient matrices to compute P : ndarray (k x k), optional Matrix such that Sigma_u = PP?, defaults to Cholesky descomp Returns: coefs : ndarray (maxn x k x k)

statsmodels.tsa.vector_ar.var_model.VARProcess.plotsim VARProcess.plotsim(steps=1000) [source] Plot a simulation from the VAR(p) process for the desired number of steps

statsmodels.tsa.vector_ar.var_model.VARProcess.mse VARProcess.mse(steps) [source] Compute theoretical forecast error variance matrices Parameters: steps : int Number of steps ahead Returns: forc_covs : ndarray (steps x neqs x neqs) Notes

statsmodels.tsa.vector_ar.var_model.VARProcess.mean VARProcess.mean() [source] Mean of stable process Lutkepohl eq. 2.1.23

statsmodels.tsa.vector_ar.var_model.VARProcess.ma_rep VARProcess.ma_rep(maxn=10) [source] Compute MA() coefficient matrices Parameters: maxn : int Number of coefficient matrices to compute Returns: coefs : ndarray (maxn x k x k)