The present code is a Matlab function for wide-sense stationarity estimation of a given signal using a novel procedure. According the theory, the objects of estimation are: the mean, the variance and the autocovariance of the signal under test. The three parameters must be time-independent in order to mark a signal as a non-stationary one.
The function provides a computation of three Boolean flags for:
1) stationarity about the mean or linear trend;
2) stationarity about the variance (and hence about the RMS-value);
3) time-invariance of the autocovariance (and hence of the autocorrelation and PSD).
An example is given in order to clarify the usage of the function. For convenience, the input and output arguments are given in the beginning of the function.
The codes are based on the theory described in:
 J. Jan. Digital Signal Filtering, Analysis and Restoration. Stevenage, The Institution of Engineering and Technology, 2000.
 D. Manolakis, V. Ingle. Applied Digital Signal Processing. Cambridge, Cambridge University Press, 2011.
 F. Wilcoxon. Individual comparisons by ranking methods. Biometrics Bulletin, Vol. 1, No. 6, pp. 80-83, Dec. 1945.
 M. Brown, B. Forsythe. Robust tests for the equality of variances. Journal of the American Statistical Association, Vol. 69, pp. 364-367, 1974.
Hristo Zhivomirov (2020). Stationarity Estimation of a Signal with Matlab (https://www.mathworks.com/matlabcentral/fileexchange/75118-stationarity-estimation-of-a-signal-with-matlab), MATLAB Central File Exchange. Retrieved .
Inspired by: Matrix Visualization with Matlab