Parametric and Nonparametric Empirical Regression Models: Case Study of Copper Bromide Laser Generation

Table 4

Results of the cross validation of the models utilizing independent data sets (70% training set and 30% evaluation set).

Model	Training set 70%			Evaluation set 30%		Number of predictors
	R2	R2 adj.	MARS	R2	R2 adj.
			GCV
			R2
MLR 0th-order, no interactions	0.945	0.942	—	0.948	0.947	3
MLR 1st-order interactions	0.948	0.945	—	0.955	0.954	6
MLR 2nd-order interactions	0.968	0.964	—	0.967	0.965	7
MARS 0th-order, no interactions	0.963	0.960	0.952	0.966	0.965	3
MARS 1st-order interactions	0.977	0.974	0.951	0.957	0.956	6
MARS 2nd-order interactions	0.965	0.962	0.953	0.966	0.965	4