An Empirical Study of Machine Learning Algorithms for Stock Daily Trading Strategy

Table 14

Trading performance of different trading strategies in CSICS. Best performance of all trading strategies is in boldface.

	Index	BAH	MLP	DBN	SAE	RNN	LSTM	GRU	CART	NB	RF	LR	SVM	XGB
AR	—	—	0.5175	0.5167	0.5163	0.5030	0.4993	0.4993	0.5052	0.5084	0.5090	0.5084	0.5112	0.5087
PR	—	—	0.7548	0.7436	0.7439	0.5414	0.4964	0.4956	0.5022	0.5109	0.5128	0.4967	0.5695	0.5026
RR	—	—	0.5252	0.5250	0.5248	0.5234	0.5224	0.5223	0.5279	0.5307	0.5311	0.5318	0.5295	0.5315
F1	—	—	0.6150	0.6108	0.6108	0.5320	0.5086	0.5082	0.5143	0.5192	0.5214	0.5132	0.5483	0.5164
AUC	—	—	0.5027	0.5024	0.5020	0.5006	0.4995	0.4996	0.5049	0.5078	0.5082	0.5086	0.5074	0.5085
WR	0.5222	0.5090	0.5559	0.5565	0.5564	0.5681	0.5720	0.5717	0.5153	0.5317	0.5785	0.5809	0.5716	0.5803
ARR	0.0633	0.2224	0.5731	0.5704	0.5678	0.5248	0.5165	0.5113	0.5534	0.6125	0.4842	0.5095	0.5004	0.4938
ASR	0.2625	0.4612	1.4031	1.4006	1.3935	1.4880	1.5422	1.5505	1.2232	1.1122	1.4379	1.5582	1.4231	1.4698
MDD	0.4808	0.6697	0.6082	0.6086	0.6130	0.5648	0.5456	0.5429	0.5694	0.7469	0.5695	0.5410	0.5775	0.5632