statsmodels.tsa.stattools.pacf_burg

statsmodels.tsa.stattools.pacf_burg(x, nlags=None, demean=True)[source]

Burg’s partial autocorrelation estimator

Parameters:

x : array-like

Observations of time series for which pacf is calculated

nlags : int, optional

Number of lags to compute the partial autocorrelations. If omitted, uses the smaller of 10(log10(nobs)) or nobs - 1

demean : bool, optional

Returns:

pacf : ndarray

Partial autocorrelations for lags 0, 1, …, nlag

sigma2 : ndarray

Residual variance estimates where the value in position m is the residual variance in an AR model that includes m lags

References

[*]Brockwell, P.J. and Davis, R.A., 2016. Introduction to time series and forecasting. Springer.