MATLAB: Linear regression model with fitlm

fitlmlinear regression modelpearsonpvalueweighted correlation

I have two arrays and I am doing a weighted correlation with the function fitlm.
If I write:
tbl = table(ones(9,1),a(:),b(:),'VariableNames',{'Weight','array1','array2'});
correlation = fitlm(tbl)
I get:
correlation =
Linear regression model: map2 ~ 1 + Weight + map1
Estimated Coefficients: Estimate SE tStat pValue ______ _____ ______ ______
    (Intercept)     0.66696    0.24971       2.671    0.036979
    Weight                0          0         NaN         NaN
    map1           -0.22041    0.39988    -0.55119     0.60141
Number of observations: 9, Error degrees of freedom: 7 Root Mean Squared Error: 0.292 R-squared: 0.0416, Adjusted R-Squared -0.0953 F-statistic vs. constant model: 0.304, p-value = 0.599
In correlation I can find almost all the values printed in the workspace, with the exeption of the p-value = 0.599
Why? Where is it and what is it?
Thank you.

Best Answer

You may have to do a separate anova call to get it:
Anova = anova(correlation);
AnovaP = Anova.pValue(2);
That works for your model.
(I usually am interested in the coefficient statistics, that are generally easier to recover.)
Related Question