I have scoured the internet and read the docs looking for an answer to this problem to no avail. I just want to be able to use the SAS TTEST proc for a two-sample T-Test but where the null hypothesis being that the difference between the two means is less than or equal to a certain value.

Every single example I’ve come upon so far merely tests again h0=0.

What I need is the logical equivalent of the pseudocode inside square brackets for the null hypothesis in the following example:

PROC TTEST DATA=MYDATA sides=2 alpha=0.05 h0=[absVal(XBar1 - XBar2) <= 100];
    TITLE "My t-test";
    CLASS LOCATION; 
    VAR TEMP;
RUN;  

If I guess and set h0 = 100 I get p-values that lead me to believe this is not what I mean.