autocorrelation

autocorrelation(x, max_lag=None, type='correlation', confidence_level=0.95)

Autocorrelation, autocovariance, or partial autocorrelation function.

Mirrors the C# Autocorrelation class. Computes the requested function out to max_lag (default floor(min(10*log10(N), N-1)), matching the C# default) and attaches the asymptotic confidence band for ACF/PACF rho values at confidence_level.

Parameters

Name Type Description Default
x array_like At least two elements. required
max_lag int Maximum lag. None (the default) uses the C# default rule. None
type ('correlation', 'covariance', 'partial') "correlation"
confidence_level float 0.95

Returns

Name Type Description
dict Keys lag, value (both :class:numpy.ndarray), type, and ci (a dict with lower/upper).

Examples

>>> from corehydropy import autocorrelation
>>> x = [5, 6, 4, 7, 3, 8, 2, 9, 1, 10, 5, 6, 4, 7, 3, 8, 2, 9, 1, 10]
>>> round(float(autocorrelation(x, max_lag=5)["value"][0]), 12)
1.0