Search code examples
wolfram-mathematicamathematica-8

How can I test for bivariate normality using Shapiro-Wilk test in Mathematica?

I've computed the Pearson correlation coefficient between two lists and am now trying to verify whether these two data sets are normally distributed in order to ascertain the validity of the the correlation.

From What I've read in the documentation, I should be able to use multivariate data with The ShapiroWilkTest function, but so far I have been unable to obtain an output. I do not receive an error, but the process hangs seemingly indefinitely. I am, however, able to obtain an output with univariate Shapiro-Wilk tests of either my dependent or independent variable almost instantly.

Here's what I've tried:

data1 = {x1,x2,x3,x4,x5,x6,x7}
data2 = {y1,y2,y3,y4,y5,y6,y7}
ShapiroWilkTest[{data1,data2}]
Solution

  • Have you by chance tried:

    ShapiroWilkTest[Transpose@{data1,data2}]

    I don't have that function, so I have no idea of the result.