pyhctsa.operations.correlation.embed2¶

pyhctsa.operations.correlation.embed2(y, tau='tau')¶

Statistics of the time series in a 2-dimensional embedding space.

Embeds the (z-scored) time series in a two-dimensional time-delay embedding space with a given time-delay, tau, and outputs a set of statistics about the structure in this space, including angular distribution, stationarity, Euclidean distances from the origin, and statistics on outliers.

Parameters:¶

y : array-like¶

The input time series (will be converted to a column vector).

tau : int or str, optional¶

The time-delay. If ‘tau’, it will be set to the first zero-crossing of the autocorrelation function (ACF).

Returns:¶

Dictionary containing:

• Distribution of angles between successive points in the embedding space.

• Stationarity of the angular distribution (across segments).

• Euclidean distances from the origin (mean, std, etc.).

• Statistics on outliers in the embedding space (e.g., area ratios).

Return type:¶

dict