hctsa results for the radial component of the supercritical Hopf normal form

This .mat file contains hctsa (v0.98) feature values for simulated time series from the radial component of the supercritical Hopf normal form with additive white Gaussian measurement noise. These data support the main analysis conducted in the paper "Tracking the distance to criticality in systems with unknown noise". The accompanying code repository can be found at Criticality.We simulated 10100 time series using all combinations of 101 equally spaced values for the control parameter (-1, -0.99, ..., 0), and 100 equally spaced values of the noise strength (0.01, 0.02, ..., 1). Time series were simulated using the Euler--Maruyama method over 1000s with a time-step of 0.001s. To avoid the effects of transient dynamics we set an initial radius of 0, coinciding with the stable fixed point of the system, and discarded the first 500s of the integration period. Finally, we down-sampled the remaining 500s to a sampling period of 0.1s, yielding 5000-sample time series. We then computed the values of 7873 hctsa features (alongside some custom features) for each time series on a distributed cluster and computed correlations of each feature to the control parameter, producing the output given in time_series_data.mat