RESAMPLING CHAOTIC TIME-SERIES

A method for resampling time series generated by a deterministic chaot ic data generating process (DGP) is proposed. Given an observed time s eries, this method potentially allows one to obtain an arbitrary numbe r of time series of arbitrary length which can be considered as a prod uct of the same unknown DGP. The notion of shadowing and brittleness o f the pseudo-orbit proves to be particularly useful in characterizing the conditions for a correct resampling. A simple practical applicatio n of the method is shown.