Temporal Features

NameDescription# Features
nc_temporal_ac The autocorrelation of the timeseries at lags one through five 5
nc_temporal_bins The power of the timeseries collected in 5 different bins between 0.008 and .25 Hz 5
nc_temporal_dynrange Dynamic range is the difference in power value (or count) between the peak and the minimum power to the right or left. Dynamic fft high and low is the difference in power X count at the peak vs left and right lowest frequencies 6
nc_temporal_entropy the entropy of the timeseries 1
nc_temporal_highfreqnoise Percentage of high frequency noise above specified index 1
nc_temporal_kurtosis kurtosis is a measure of how outlier prone a distribution is. Distributions that are more outlier-prone than normal have kurtosis greater than 3. Distributions that are less outlier-prone than normal have kurtosis less than 3. 1
nc_temporal_mean The mean response over time 1
nc_temporal_peaks Calculations for minimum and maximum peak values, as well as amplitude 9
nc_temporal_psd Power spectrum density 83
nc_temporal_skewness skewness of timecourse 1
nc_temporal_energyratio We calculate the spectral power below 0.10 Hz over the power between 0.15 and 0.25 Hz 1