Skip to main content

Table 12 Results of McNemar’s test on allstock_VAE trained models

From: An efficient stock market prediction model using hybrid feature reduction method based on variational autoencoders and recursive feature elimination

Stocks

LSTM (with attention) versus LSTM basic

LSTM (with attention) versus light GBM

LSTM (with attention) versus SVM

Score

p value

Score

p value

Score

p value

ALBRK

62

1

80

0.698

72

0.683

AKBNK

92

4.77E−06

102

0

174

0

HALKB

175

4.03E−09

164

0.066

104

0.286

ISCTR

18

0.064

66

0.004

106

0

SKBNK

41

0.594

71

0.742

77

0.001

TSKB

108

0.002

119

0.702

129

0.365

VAKBN

131

1.16E−06

159

0

190

0

YKBNK

150

2.80E−10

184

0.001

71

0.204

  1. Bold numbers indicate the best performance