statsmodels.tsa.stattools.ccf

statsmodels.tsa.stattools.ccf(x, y, unbiased=True)[source]

cross-correlation function for 1d

Parameters:
  • y (x,) – time series data
  • unbiased (boolean) – if True, then denominators for autocovariance is n-k, otherwise n
Returns:

ccf – cross-correlation function of x and y

Return type:

array

Notes

This is based np.correlate which does full convolution. For very long time series it is recommended to use fft convolution instead.

If unbiased is true, the denominator for the autocovariance is adjusted but the autocorrelation is not an unbiased estimtor.