Kolmogorov-Smirnov test for inverse Gaussian distribution

Hello ... I has data and I want to use a goodness-of-fit test (Kolmogorov-Smirnov) to indicate whether the data follow the inverse Gaussian distribution or not ... can you help me, how do I run this test ??

[Jeff Miller]

Are the parameters of the inverse Gaussian distribution specified theoretically in advance, or do you want to estimate them from the data?

[Abas Arkawazi]

yes I want to estimate them from the data.

[Jeff Miller]

Sorry, I don't have an answer for that case.

Answer 1

[Jakob B. Nielsen]

The Kolmogorov-Smirnov test checks if your data is from a standard normal distribution, not an inverse gaussian. Of course, it is always good to check, first, if your data is from a standard normal distribution, before looking for alternative distributions :)

Simply [h,p]=kstest(data) will give you that info.

In case the null hypothesis is rejected, there is a code on the file exchange for testing for the inverse gaussian distribution, that you might check out. https://se.mathworks.com/matlabcentral/fileexchange/24036-anderson-darling-goodness-of-fit-test-to-inverse-gaussian-distbtn?focused=5114587&tab=function

[Adam Danz]

Note that there is a modified Kolmogorov-Smirnov test for the inverse gaussian. Whether there exists Matlab code that implements this, I do not know.