kedm.simplex

kedm.simplex(lib: numpy.ndarray[numpy.float32], pred: numpy.ndarray[numpy.float32], *, target: numpy.ndarray[numpy.float32] = None, E: int = 1, tau: int = 1, Tp: int = 1) → numpy.ndarray[numpy.float32]

Predict a time series from another using Simplex projection.

Parameters:

  • lib – Library time series

  • pred – Prediction time series

  • target – Target time series (defaults to lib if None)

  • E – Embedding dimension

  • tau – Time delay

  • Tp – Prediction interval

Returns:

Prediction result

Note

If both lib and pred are 2D arrays, mixed multivariate embedding is peformed, where each time series is embedded into an E-dimensional state space.

Examples

Forecast:

>>> kedm.simplex(x[:100], x[100:200], E=2, Tp=1)

Cross mapping:

>>> kedm.simplex(x, y, target=y, E=3, Tp=0)

Multivariate forecast:

>>> kedm.simplex(xs, ys, target=y, E=4, Tp=1)