TimeSeriesSRC.xcorr¶

TimeSeriesSRC.xcorr(a, b, maxlag=20, flag='none')¶

Compute the cross-correlation (or autocorrelation) between two sequences.

Produces a symmetric array of length 2*maxlag + 1 covering lags -maxlag to +maxlag.

Parameters:

a (array-like) – First 1-D sequence (length N).
b (array-like) – Second 1-D sequence (same length as a).
maxlag (int, optional) – Maximum lag to compute; must be less than N. Default 20.
flag ({'biased', 'unbiased', 'coeff', 'none'}, optional) –
Normalization method:
- 'biased' — divide by N (biased estimate).
- 'unbiased' — divide by N - |k| (unbiased estimate).
- 'coeff' — normalize so that the zero-lag value is 1.0.
- 'none' — no scaling (raw inner product). Default.

Returns:

c – Cross-correlation values at lags -maxlag, ..., 0, ..., +maxlag. The zero-lag value is at index maxlag.

Return type:

ndarray, shape (1, 2*maxlag+1)

Examples

>>> import numpy as np
>>> from scipy.signal import lfilter
>>> from TimeSeriesSRC.basefunctions.xcorr import func_xcorr
>>> e = np.random.default_rng(0).standard_normal(500)
>>> y = lfilter([1], [1, -0.8], e)
>>> acf = func_xcorr(y, y, maxlag=20, flag='biased')
>>> acf.shape
(1, 41)