pyhctsa.operations.correlation.compare_min_ami¶

pyhctsa.operations.correlation.compare_min_ami(y, bin_method='std1', num_bins=10)¶

Assess the variability in the first minimum of automutual information (AMI) across binning strategies.

This function computes the first minimum of the automutual information function for a time series using various histogram binning strategies and numbers of bins. It summarizes how the location of the first minimum varies across these different coarse-grainings. The first minimum is taken to be the lag at the first change in the sign of the slope of the AMI, so it is a local maximum if the AMI first rises. If the slope never changes sign, the maximum lag considered is used.

Parameters:¶

y : array-like¶

The input time series.

bin_method : str, optional¶

The method for estimating mutual information (passed to histogram_ami). Default is ‘std1’.

num_bins : int or array-like, optional¶

The number of bins (or list of bin counts) to use for AMI estimation. Default is 10.

Returns:¶

Dictionary containing statistics on the set of first minimums of the automutual information function. nprompeaks is the number of prominent peaks (local maxima) of the first-minimum lag as a function of the number of bins: peaks that rise at least 5% of the range of the lags above the higher of the valleys on either side of them, so a small fluctuation does not add a peak.

Return type:¶

dict