Validation is a necessary step for the model application [31]. The
validation of a model is that it generates good testable hypotheses
relevant to important problems. We should test these models in the
following way. Three commonly used indexes, including the total
deviation sum of squares (TSS), regression sum of squares (RSS) and
error sum of squares (ESS), are mainly considered here for evaluating
the fitting efficiency (R2
) of these methods. Given a linear regression
model: