Skip to main content

Table 3 Hyperparameter values determined by Bayesian optimization

From: Forecasting VaR and ES by using deep quantile regression, GANs-based scenario generation, and heterogeneous market hypothesis

 

FTSE100

N225

SPX500

DAX

S 1

S 2

S 3

S 4

S 1

S 2

S 3

S 4

S 1

S 2

S 3

S 4

S 1

S 2

S 3

S 4

\(\tau = 0.01\)

                

 \(N_{f}\)

2

2

1

1

2

2

2

1

2

1

1

1

1

1

1

1

 \(N_{h}\)

80

48

98

43

89

71

45

100

70

94

83

67

35

61

79

39

 \(r\)

2

6

4

2

3

8

1

1

3

2

10

6

8

2

2

4

 \(N_{L}\)

1

3

3

4

3

3

4

4

1

2

4

4

3

4

2

3

 \(LT\)

1

0

1

0

0

0

0

0

0

1

1

0

0

0

0

0

 \(Epoch\)

205

373

108

346

218

178

165

356

438

347

434

157

273

152

228

90

\(\tau = 0.05\)

                

 \(N_{f}\)

4

2

2

1

3

1

1

1

2

2

1

1

2

1

1

1

 \(N_{h}\)

21

81

60

88

47

44

24

63

42

72

70

8

100

100

100

24

 \(r\)

6

3

4

8

10

4

1

5

10

2

9

6

1

1

3

8

 \(N_{L}\)

3

3

1

3

3

2

1

4

1

2

3

3

4

4

2

3

 \(LT\)

1

0

1

0

0

0

1

0

0

1

0

0

1

1

1

1

 \(Epoch\)

300

434

162

456

248

115

122

330

133

119

137

500

155

500

81

269

  1. In this paper, the model selection is realized by different values of \({\text{LT}}\) and \({\text{r}}\). S refers to sub-series, which is aggregated by decomposed IMFs according to fuzzy entropy and approximation criterion.