Equivalence testing for non parametric data in SAS?

[ChrilleBob]

Hi!

I have this problem nagging me and I would love some input!

I got this set of data with measurements made with two different manufacturing methods which I would like to compare with each other, and test for equivalence between the methods. My first thought was to conduct a regular TOST with proc ttest but I discarded that idea when i reviewed the data 🙂 The data consists of 15 observations for each manufacturing method, and I cannot assume normality due to the distribution of the data. The data is continuous (not ordinal nor nominal).

My question: Is there anybody here that has experience with non-parametric equivalence testing? Preferably in SAS Base (or Enterprise Guide, Excel can work but I rather stay away from it).
I've made some googeling and found some results for a Wilcoxon-Mann-Whitney TOST but no real connection to SAS.

Thank you very much for your time!

/Christian B.

[SteveDenham]

I really don't know if you can extend the methods to TOST but I would look at PROC NPAR1WAY for Hodges-Lehman estimates of location shift, and perhaps to PROC MULTTEST for bootstrap resampling of the t-test.

Good luck with this.

Steve Denham

[ChrilleBob]

Thank you Steve, much appreciated! I will look into this 🙂

/Christian.

[charles77]

Hi ChrilleBob,
Did you find a way to make Mann whitney Wilcoxon TOST using SAS ?

Charles.