The codes belong to the following article:
Farokhzadi, M., Hossein-Zadeh, G.A., Soltanian-Zadeh, H., 2018. Nonlinear Effective Connectivity Measure Based on Adaptive Neuro Fuzzy Inference System and Granger Causality. J Neuroimage 181, 382–394. https://doi.org/10.1016/j.neuroimage.201...
The shared MATLAB codes contain:
1) Generating three time series with 4000 samples (n=1,…,4000) based on the following equations:
x_1 (n)=3.4x_1 (n-1)(1-x_1^2 (n-1)) e^(-x_1^2 (n-1) )+ε_1 (n)
x_2 (n)=3.4x_2 (n-1)(1-x_2^2 (n-1)) e^(-x_2^2 (n-1) )+cx_2 (n-1) x_1 (n-1)+ε_2 (n)
x_3 (n)=3.4x_3 (n-1)(1-x_3^2 (n-1)) e^(-x_3^2 (n-1) )+0.3x_2 (n-1)+0.5x_1^2 (n-1)+ε_3 (n)
The signal-to-noise ratio (SNR) is 0 dB.
The parameter c, which indicates the coupling strength of (x1→x2) connection, is set to 0.5.
2) Implementation of ANFISGC criterion to detect both linear and nonlinear causal connections among brain regions.
In the Matlab functions, the inputs and outputs are described.