statsmodels.tsa.stattools.ccovf¶
-
statsmodels.tsa.stattools.
ccovf
(x, y, unbiased=True, demean=True)[source]¶ crosscovariance for 1D
Parameters: - y (x,) – time series data
- unbiased (boolean) – if True, then denominators is n-k, otherwise n
Returns: ccovf – autocovariance function
Return type: array
Notes
This uses np.correlate which does full convolution. For very long time series it is recommended to use fft convolution instead.