COVTEST <output_line> / SAMPLES=<data1>,<data2,... tests the equality of covariance matrices of <data1>,<data2>,... by assuming multivariate normality. See e.g. "Methods of Multivariate Statistics" by M.S.Srivastava (2002). The testing procedure is also continued by a randomization test using the same test statistics but without assumption of multivariate normality. The number of replicates in the simulation process is given by a SIMUMAX specification. Default is SIMUMAX=10000. (An example in T2TEST?) M = Comparing multivariate samples