`xcov`

Cross-covariance of two sequences, or autocovariance of one.

Category: Data Import and Analysis > Descriptive Statistics

Syntax

[c, lags] = xcov(x)

Autocovariance of x at lags from -(N-1) to N-1.

[c, lags] = xcov(x, y)

Cross-covariance of x and y.

Description

xcov is xcorr applied after the mean of each sequence has been subtracted. Removing the mean keeps a constant offset from dominating the result, so the output reflects how the fluctuations around the average relate to each other at each lag.

The autocovariance at lag 0 is N times the variance (with divisor N).

Input arguments

x, y: Numeric vectors.

Output arguments

c: Covariance values, one per lag.
lags: Lag indices.

Example

Autocovariance

xcov([1 2 3])

Output

ans = 1×5

    -1     0     2     0    -1

Notes

xcov belongs to the Signal Processing Toolbox in MATLAB.

xcov