statsmodels.tsa.stattools.levinson_durbin_pacf

statsmodels.tsa.stattools.levinson_durbin_pacf(pacf, nlags=None)[source]

Levinson-Durbin algorithm that returns the acf and ar coefficients.

Parameters:
pacfarray_like

Partial autocorrelation array for lags 0, 1, … p.

nlagsint, optional

Number of lags in the AR model. If omitted, returns coefficients from an AR(p) and the first p autocorrelations.

Returns:
arcoefsndarray

AR coefficients computed from the partial autocorrelations.

acfndarray

The acf computed from the partial autocorrelations. Array returned contains the autocorrelations corresponding to lags 0, 1, …, p.

References

[1]

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