Skip to main content

Table 5 Gaps between actual and predicted returns under the linear covariance function

From: Predicting the returns of the US real estate investment trust market: evidence from the group method of data handling neural network

Day

Linear-cov-30/30

Linear-cov-100/30

Linear-cov-200/30

Linear-cov-300/30

1

− 1.01

− 0.95

− 0.57

− 0.62

2

1.62

− 0.29

0.19

− 0.32

3

− 0.29

0.28

0.48

0.25

4

− 0.10

0.48

0.48

0.09

5

− 0.54

0.82

0.53

0.07

6

0.70

0.14

0.42

− 0.60

7

− 1.11

1.53

0.85

0.24

8

− 1.94

0.47

0.18

− 0.64

9

− 1.87

1.08

0.76

− 0.22

10

− 2.06

1.21

0.59

− 0.56

11

− 1.20

2.14

1.61

0.28

12

− 2.80

0.91

0.89

− 0.55

13

− 4.33

− 1.84

− 0.40

− 2.12

14

− 5.81

− 1.67

0.11

− 0.76

15

− 11.85

0.21

0.81

0.25

16

− 9.40

0.56

0.87

0.46

17

− 8.42

− 0.88

− 0.44

− 1.86

18

− 17.04

− 0.70

− 0.21

− 2.14

19

− 29.16

− 0.71

− 0.34

− 1.06

20

8.07

0.38

0.72

− 0.48

21

− 11.14

− 0.19

0.07

− 2.87

22

− 5.71

1.35

1.60

− 0.93

23

− 5.57

1.01

1.23

− 0.02

24

− 4.74

0.49

0.28

− 0.61

25

− 2.10

1.86

0.91

0.46

26

− 2.63

− 1.74

− 2.00

− 2.11

27

− 2.04

0.89

− 0.17

− 0.76

28

− 2.86

− 0.15

− 0.41

− 0.97

29

− 2.03

0.37

0.03

− 0.25

30

− 2.63

− 1.35

− 0.68

− 1.33

31

− 1.30

0.17

0.51

− 0.46

32

0.31

− 0.19

1.86

0.31

MSE

57.22

1.05

0.69

1.08