Software - Novel Surrogate Hypothesis Tests

This page contains software for Matlab for generating surrogates, calculating calibrated time-delayed mutual information, and calculating linear mutual information for performing the novel surrogate data tests described in reference [1]. Download the following code:

surrogates.m	Generates constrained realisation surrogate versions of a given signal using the IAAFT method of Schreiber and Schmitz [2].
order_mutual.c	Calculates estimates of the time-delayed mutual information using Newton-Cotes numerical integration. Written in C with Matlab MEX integration.
order_mutual.dll	Above code compiled as a DLL for direct use with Matlab under Microsoft Windows 32. Place the DLL in a directory accessible to Matlab and invoke as with any other function.
order_mutual.m	Useage instructions for above routine.
linear_mutual.m	Calculates linear time-delayed mutual information.
calibrate.m	Calculates baseline time-delayed mutual information for subsequent calibration of further time-delayed mutual information calculations using mutual_calib.m.
mutual_calib.m	Calculates calibrated time-delayed mutual information.

Typing 'help (function name)' shows instructions for each routine. This ZIP archive contains the speech data described in the paper.

[1] M.A. Little, P.E. McSharry, I.M. Moroz, S.J. Roberts (2006), Testing the assumptions of linear prediction analysis in normal vowels. Journal of the Acoustical Society of America, 119(1): pp. 549-558.

[2] T. Schreiber, A. Schmitz (2000), Surrogate time series. Physica D-Nonlinear Phenomena, 142(3-4): pp. 346-382.