Forecast of complex financial big data using model tree optimized by bilevel evolution strategy

Journal of Big Data

Table 4 Accuracy comparison of regression problems

	Air quality		Energy predict		Electrical stability		Gas sensor		Election result
	Train	Test	Train	Test	Train	Test	Train	Test	Train	Test
Linear regression	0.49	0.45	0.16	0.16	0.65	0.64	0.58	0.59	0.98	0.98
Lasso regression	0.47	0.46	0.15	0.15	0.08	0.08	0.58	0.58	0.98	0.98
Partial minimum error	0.48	0.45	0.15	0.14	0.65	0.64	0.56	0.56	0.98	0.98
Neural network	0.65	0.61	0.57	0.30	0.87	0.85	0.67	0.61	0.60	0.60
XG boost	0.78	0.58	0.74	0.38	0.97	0.93	0.84	0.84	1.00	1.00
Bilevel GA by related work	0.50	0.49	0.59	0.48	0.72	0.60	0.70	0.64	0.95	0.93
Bilevel GA by this study	0.79	0.69	0.76	0.61	0.98	0.95	0.91	0.89	1.00	1.00