Closed finmod closed 5 years ago
Thanks. If your suggestion is to add support for the analysis of trajectories, I think that this is already done, unless there is something more implicit in the idea (I confess that I have not yet read the paper in depth).
The functions embed
and reconstruct
(now imported from DelayEmbeddings) can be applied to multidimensional variables. The creation and analysis of recurrence plots do not "care" if the data set is reconstructed from a one-dimensional time series, or if it is a trajectory in origin.
Yeap, all RQA functions can accept a Dataset
(which is a trajectory or an embedded timeseries). I'll close this for now, will re-open if we have misunderstood.
Excellent. This is a strength of Dataset and of the new DelayEmbeddings that should be emphasized in the docs and in the examples by showing that RecurrenceAnalysis is multidimensional.
@heliosdrm Just wanted to bring to your attention this extension to trajectories (in the DynamicalSystems sense) i.e. multidimensional timeseries of RecurrenceAnalysis. (https://www.frontiersin.org/articles/10.3389/fpsyg.2016.01835/full)
The illustration with Lorenz63 demonstrates the usefulness of the trajectory approach versus the timeseries approach.