Skip to main content

Table 1 Simulation experiment results: bias

From: Association of longitudinal pet ownership with wheezing in 3-year-old children using the distributed lag model: the Japan Environment and Children’s Study

time

γ=0

γ=0.975

γ=0.9999

Single

Multi

DLM

Single

Multi

DLM

Single

Multi

DLM

Scenario 1

 6

0.0

0.1

0.0

93.2

-0.2

0.1

131.7

1.4

-0.8

 12

-0.3

-0.2

-0.1

97.3

0.5

-0.1

131.9

-3.1

-0.5

 18

-0.5

-0.4

-0.1

100.1

-0.4

-0.1

132.1

0.2

-0.2

 24

0.1

0.2

-0.1

101.9

0.0

-0.1

132.3

-0.1

0.1

 30

-0.3

-0.2

-0.1

102.8

0.2

-0.1

132.3

2.1

0.3

 36

-0.2

-0.1

-0.1

102.7

-1.6

-0.1

132.3

0.0

0.4

 42

-0.3

-0.3

-0.1

102.2

1.1

0.0

132.3

0.6

0.3

 48

-0.1

0.0

-0.1

100.3

-0.1

0.1

132.1

-1.7

0.2

 54

-0.1

-0.1

-0.1

97.6

1.1

0.2

131.9

0.0

0.0

 60

-0.3

-0.2

-0.1

93.4

-0.5

0.3

131.7

0.0

-0.4

Scenario 2

 6

0.4

0.3

0.1

98.9

-0.2

-0.1

139.4

-1.2

-0.1

 12

-0.3

-0.2

-0.2

99.1

0.1

-0.1

134.8

3.4

-0.2

 18

-0.4

-0.3

-0.2

99.7

-0.1

0.0

131.3

-0.6

-0.2

 24

-0.3

-0.2

-0.2

100.3

0.4

0.1

129.0

0.5

-0.1

 30

-0.3

-0.2

-0.1

100.6

-0.1

0.2

127.7

-3.1

-0.1

 36

0.0

0.1

0.0

100.7

1.1

0.2

127.6

-2.5

0.0

 42

-0.3

-0.2

0.0

100.1

-0.8

0.1

128.7

3.0

0.0

 48

-0.2

-0.1

0.0

99.4

-0.3

0.0

130.8

2.3

0.1

 54

-0.5

-0.4

0.0

98.7

-0.3

-0.2

134.1

-3.2

0.2

 60

-0.1

0.0

0.0

98.3

-0.4

-0.4

138.6

1.1

0.3

Scenario 3

 6

0.3

0.3

0.9

33.6

0.5

-0.1

49.4

0.9

-3.4

 12

0.4

0.4

-4.2

35.1

0.5

-4.1

49.5

0.3

-3.2

 18

-0.3

-0.3

-0.9

36.6

-0.9

-0.6

49.6

-1.6

0.9

 24

-0.1

-0.1

4.5

38.7

-0.1

4.6

49.8

-0.7

5.6

 30

-0.1

-0.1

9.4

41.3

0.4

9.4

50.0

2.3

9.4

 36

-6.5

-6.5

-12.3

20.8

-5.9

-12.4

25.2

-5.8

-13.2

 42

-0.4

-0.4

-11.4

21.9

-0.6

-11.6

25.2

-7.5

-12.8

 48

6.5

6.6

11.5

43.1

6.3

11.4

50.1

9.2

10.5

 54

0.2

0.2

6.2

39.9

0.1

6.3

49.9

3.4

6.5

 60

-0.5

-0.5

-2.6

37.7

0.3

-2.1

49.7

-0.3

-0.1

  1. Bias rounded to the fourth decimal place and multiplied by \({10}^{2}\)
  2. Single, the exposure at each time point is used in a separate regression model; Multi, the exposures at each time point are simultaneously entered into a single regression model; DLM, the Distributed Lag Model
  3. γ represents the correlation used to generate data between exposures
  4. Scenario 1: Exposure effects are constant regardless of time point. Scenario 2: Exposure effects’ magnitude is an upward convex quadratic curve. Scenario 3: Exposure effects exist only from \(t = 36 \text{ to } 47\)