Find P-values from regression

I have response variable matrix Y and explanatory variable matrices X1,X2,X3 and X4.I want to find the p value of the explanatory matrices. I dont use Linearmodel.fit method.I use svd and recursive least squares to find the regression coefficients.So how do I find the p values from X matrix and Y matrix.

i=1:10;

X(i)=[x1 x2 x3 x4];

I tried [r,p]= corrcoef(X) but i get same values of p for all explanatory variables

Help would be appreciated

[Roger Wohlwend]

I am not sure what you mean with p-value. When you do a regression you get p-values for the coefficients that tell you if the coefficients are significant. However, that seems not what you have in mind. What do you mean with p-value?