Hi,
You can use MATLAB's Statistics and Machine Learning Toolbox. It offers a function "fitlm" which you can use as a linear model.
% Fit using fitlm:
% Here X will be quadratic in your case
mdl = fitlm(X,y);
% mdl.Coefficients is a table with columns Estimate, SE, tStat, pValue
% Rows correspond to the terms (Intercept, x, x^2)
disp(mdl.Coefficients);
Please refer to the documentation of "fitlm" for more information:
Hope this helps!
