number of predictors in the model. The Adj. R2 increases only if
the new term improves the model performance more than what
would be expected by chance, and decreases when a predictor
improves the model less than what would be expected by chance.
The Pred. R2 measures how well the model predicts responses for
new observations. Furthermore, an adequate precision value of
89.23 that measures the signal-to-noise ratio indicated that the
established model can be used to navigate the design space bound for the %AAr data set, an F-value of 19.65 was obtained with a
corresponding P < 0.0001 value, indicating significant fit to the
quadratic model; and a non-significant (F-value ¼ 1.30, P ¼ 0.337)
lack-of-fit. The individual linear effects of heating temperature,
exposure time, and initial AA content; and the individual quadratic
effects of temperature and initial AA content significantly influenced
%AAr in the heated SFJ, resulting to the predictive model
Equation (6). The calculated model R2, Adj. R2 and Pred. R2 values of
0.803, 0.762, and 0.662, respectively, similarly indicate goodnessof-
fit and performance, and are in reasonable agreement with
each other.