Table 8.

Brusselator model structure for concentration of X obtained from forward selection

Selection order	Term index	Terms	Param’s	SSE
1	73	Xt−1Yt−1	0.4519	0.2788
2	75	Xt−1Yt−3	0.1368	0.1295
3	439	Xt−3Y2t−1	−0.0669	0.1046
4	79	Xt−2Yt−2	0.0761	0.0978
5	429	Xt−2Y2t−1	0.0847	0.0926
6	425	Xt−2X2t−3	1.4923	0.0864
7	294	B2t−1Xt−2	−0.4418	0.0816
8	11	Yt−2	−0.0073	0.0750
9	419	Xt−1Y2t−3	−0.0093	0.0716
10	43	At−3Yt−1	−0.0142	0.0695
11	147	At−1Bt−3Yt−3	0.0126	0.0666
12	402	X2t−1Yt−1	−0.5576	0.0645

The parameters P_k and model terms Θ_k are given for the case of no. of training samples = 100. The forward and two-stage methods both selected a different set of terms from the pool of 454 candidates