where yi is the measured value for the ith sample; y ¼ Pn
i ¼ 1 yi=n is
the average value of observations y1; y2; …yn; y^
i is the fitted response
values for the ith sample.
The test of goodness of fit is realized by the coefficient of
determination R2
. R2 can explain the regression curve fitting degree
of observed data. When 0oR2o1, just the closer R2 approaches 1,
the better fitting efficiency we get