Large-Scale Portfolio Optimization Using Multiobjective Evolutionary Algorithms and Preselection Methods

Table 3

indicators of various algorithms (FEs = 20000).

Algorithms	S-1			S-2
	1000 stocks	P-1 (200 stocks)	P-2 (200 stocks)	1000 stocks	P-1 (200 stocks)	P-2 (200 stocks)
NSGAII
Best	0.0045
Worst	0.0054	0.0021	0.0017
Mean	0.0051	0.0014	0.0012
Std
-test	—	1	1	—	1	1
MODE-SS
Best	0.0031
Worst	0.0044	0.0011
Mean	0.0039
Std
-test	—	1	1	—	1	1
MODE-NDS
Best	0.0012
Worst	0.003	0.0014	0.0014
Mean	0.0021	0.001
Std
-test	—	1	1	—	1	1
MOCLPSO
Best	0.0016	0.001	0.0011
Worst	0.0028	0.0018	0.0018	0.0011
Mean	0.0022	0.0014	0.0014
Std
-test	—	1	1	—	1	1
NMOEA/D
Best
Worst	0.0012	2.96E-04	3.39E-04	0.001
Mean
Std
-test	—	1	1	—	1	1