Regress function provides same results regardless of predictors

Hi all,

I am trying to calculate the linear regression (R squared, p-value) of two variables w and z, once by using variable w as predictor and then using variable z as the predictor. I have written the following script:

%a-predictor

%b-response

a=a';

b=b';

x1=ones(size(a,1),1);

X=[x1 a];

y=b;

[~,~,~,~,stats2] = regress(y, X);

R2=stats2(1);

p2=stats2(3);

reg=[R2 p2];

where I interchange variables a and b being the former firstly "w" and then "z" and the opposite for the latter. However, in either cases I get the same values for R2 and p2. When I try with "Regression Learner App" I do get different values for the two cases. Does anybody know what's going on?

Thanks in advance!

Katerina