Skip to main content

Table 2 Classification results with ANN/DNN classifiers using entire untransformed data

From: Predicting the daily return direction of the stock market using hybrid machine learning algorithms

# of hidden layers

Training (MSE)

Validation (MSE)

Testing (MSE)

Total (MSE)

10

57.3 (0.3058)

53.8 (0.3164)

57.3 (0.3124)

56.8 (0.3084)

12

57.5 (0.3055)

54.1 (0.3129)

57.3 (0.3110)

56.9 (0.3074)

14

57.8 (0.3041)

53.8 (0.3127)

57.6 (0.3075)

57.2 (0.3059)

16

58.6 (0.3034)

54.9 (0.3160)

58.1 (0.3099)

57.9 (0.3063)

18

58.2 (0.3045)

53.3 (0.3143)

58.1 (0.3095)

57.5 (0.3067)

20

59.1 (0.3052)

54.4 (0.3186)

58.4 (0.3146)

58.3 (0.3086)

22

57.7 (0.3041)

54.1 (0.3169)

58.1 (0.3099)

57.2 (0.3069)

24

57.0 (0.3071)

55.7 (0.3139)

58.1 (0.3066)

57 (0.3081)

26

55.4 (0.3144)

54.9 (0.3245)

58.1 (0.3143)

55.8 (0.3159)

28

54.6 (0.3163)

54.6 (0.3175)

58.6 (0.3074)

55.2 (0.3151)

30

54.5 (0.3163)

53.1 (0.3232)

58.1 (0.3215)

54.8 (0.3181)

35

58.3 (0.3050)

54.9 (0.3169)

57.8 (0.3121)

57.7 (0.3079)

40

56.3 (0.3354)

53.3 (0.3584)

57.0 (0.3399)

56.0 (0.3395)

45

58.0 (0.3055)

53.8 (0.3201)

57.0 (0.3113)

57.2 (0.3085)

50

58.3 (0.3034)

53.6 (0.3252)

57.3 (0.3134)

57.4 (0.3081)

100

54.5 (0.3354)

53.3 (0.3353)

57.0 (0.3219)

54.7 (0.3334)

500

55.4 (0.3474)

53.8 (0.3570)

57.3 (0.3386)

55.5 (0.3475)

1000

57.3 (0.3383)

54.1 (0.3521)

57.3 (0.3383)

56.8 (0.3404)