Skip to main content

Table 4 Classification results with ANN/DNN classifiers using transformed data with PCs = 31

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

56.1 (0.3067)

54.4 (0.3121)

58.9 (0.3095)

56.3 (0.3079)

12

61.6 (0.3030)

56.8 (0.3253)

58.4 (0.3141)

60.4 (0.3080)

14

54.6 (0.3237)

54.9 (0.3111)

58.9 (0.3051)

55.3 (0.3190)

16

61.0 (0.2980)

56.5 (0.3087)

59.4 (0.3084)

60.1 (0.3011)

18

54.9 (0.3145)

55.4 (0.3160)

59.2 (0.3091)

55.6 (0.3139)

20

55.0 (0.3096)

56.5 (0.3083)

59.7 (0.3079)

56.0 (0.3092)

22

55.6 (0.3097)

56.8 (0.3120)

59.9 (0.3059)

56.4 (0.3095)

24

54.1 (0.3105)

54.1 (0.3133)

58.9 (0.3132)

54.8 (0.3113)

26

56.9 (0.3228)

54.4 (0.3191)

58.6 (0.3125)

56.8 (0.3207)

28

57.1 (0.3049)

54.9 (0.3136)

59.4 (0.3081)

57.1 (0.3067)

30

54.8 (0.3152)

54.4 (0.3142)

58.4 (0.3085)

55.2 (0.3140)

35

58.2 (0.3049)

55.4 (0.3167)

58.9 (0.3083)

57.9 (0.3072)

40

55.3 (0.3111)

54.6 (0.3163)

58.6 (0.3071)

55.7 (0.3113)

45

59.2 (0.3003)

55.7 (0.3147)

58.1 (0.3081)

58.5 (0.3036)

50

57.9 (0.3040)

54.9 (0.3140)

58.4 (0.3070)

57.5 (0.3059)

100

58.6 (0.3044)

54.4 (0.3131)

58.9 (0.3061)

58.0 (0.3060)

500

60.4 (0.3117)

55.4 (0.3436)

58.6 (0.3233)

59.4 (0.3182)

1000

57.7 (0.3237)

56.0 (0.3405)

58.9 (0.3293)

57.6 (0.3271)