Iterated, Amplitude Adjusted Wavelet Transform (IAAWT) for time-series randomisation

Time series randomisation, preserving pointwise Holder structure and the original data values

Given a time-series, this algorithm generates random variants where the original values are all preserved (but their positions randomised) but the pointwise Holder structure is fixed. This is useful for various forms of hypothesis testing. See:
Keylock, C.J. 2017. Multifractal surrogate-data generation algorithm that preserves pointwise
Hölder regularity structure, with initial applications to turbulence, Physical Review E 95, 032123,
https://doi.org/10.1103/PhysRevE.95.032123.