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

Table 2

Results from constructed parametric and nonparametric regression models for estimation of output laser power $P_{\text{out}}$: MLR (Multiple Linear Regression) and MARS (Multivariate Adaptive Regression Splines).

Model	R2	R2 adj.	MARS GCV R2	Std. Err. of the Estimate	Number of predictors
MLR 0th order, no interactions	0.946	0.944	—	7.92540	3
MLR 1st order interactions	0.950	0.948	—	7.63382	4
MLR 2nd order interactions	0.967	0.965	—	6.27075	7
MARS 0th order, no interactions	0.965	0.963	0.952	6.50470	3
MARS 1st order interactions	0.973	0.970	0.951	5.82838	5
MARS 2nd order interactions	0.972	0.970	0.953	5.89545	6