pyhctsa.operations.distribution.remove_points¶

pyhctsa.operations.distribution.remove_points(y, remove_how='absfar', p=0.1, remove_or_saturate='remove')¶

How time-series properties change as points are removed.

Removes a proportion, p, of points from the time series according to a specified rule, and computes a set of statistics before and after the change.

Parameters:¶

y : array-like¶

The input time series.

remove_how : {'absclose', 'absfar' (default), 'min', 'max', 'random'}, optional¶

How to remove points from the time series:

• ’absclose’: those that are the closest to the mean,

• ’absfar’: those that are the furthest from the mean (default),

• ’min’: the lowest values,

• ’max’: the highest values,

• ’random’: at random.

p : float, optional¶

The proportion of points to remove (default: 0.1).

remove_or_saturate : {'remove', 'saturate'}, optional¶

Whether to remove points (‘remove’) or saturate their values (‘saturate’). Default is ‘remove’.

Returns:¶

Statistics including the change in autocorrelation, time scales, mean.

Return type:¶

dict