File Exchange

image thumbnail

Non-Gaussian process generation

version (29.7 KB) by E. Cheynet
A non-Gaussian random process is generated from a Gaussian-distributed white noise


Updated 19 Jul 2015

View License

In the present files, the method to transform of a Gaussian-process into a non-Gaussian one is based on the moment-based Hermite transformation model (MBHTM), and uses a cubic transform. It has been described in [1], but I relies mainly on [2] for the implementation of the code. The non-Gaussianity is introduced by a target skewness and a target kurtosis. However, the transformation works only for a limited range for the skewness and kurtosis (see [2] for more details).
3 .m files are included:
- MBHTM.m which is the main function to generare the non-Gaussian process
- Example.m which is the example file
- fitDistEtienne.m which is used in the Example.m file. it is inspired from the matlab function fitdist.
This is the first version of the script, and therefore, some changes are excpected soon. I did not carry out anything new. All the credits goes to [1] and [2]. Any comment or proposition to improve the script is warmly welcomed !
[1] Gurley, K. R., Tognarelli, M. A., & Kareem, A. (1997). Analysis and simulation tools for wind engineering. Probabilistic Engineering Mechanics, 12(1), 9-31.
[2] Denoël, V. (2005). Application des méthodes d'analyse stochastique à l’étude des effets du vent sur les structures du génie civil. Unpublished doctoral thesis, University of Liège. (in French, p. 229)

Cite As

E. Cheynet (2020). Non-Gaussian process generation (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (1)

Sevenbo Tan

Very helpful. However, thesis is in French.





-typo again


MATLAB Release Compatibility
Created with R2013a
Compatible with any release
Platform Compatibility
Windows macOS Linux