Skip to main content

Table 3 Model results of 2015 dengue epidemic using the matrix of population mobility as spatial weights

From: Evaluating neighborhood structures for modeling intercity diffusion of large-scale dengue epidemics

Independent variables

Model 1

Model 2

Model 3

Intercept

− 3.08***(0.90)

− 1.56(0.81)

− 0.53(0.92)

Spatial-lag of dengue incidence

 1st order neighbors in pre-epidemic period ρt1

0.50***(0.09)

0.27*(0.08)

− 0.007(0.16)

 1st order neighbors in epidemic period ρt2

0.47***(0.08)

0.467***(0.08)

 2nd order neighbors in pre-epidemic period γt1

0.40*(0.19)

Urbanization levelsa

 Rural area β1

− 0.11(0.25)

− 0.10(0.21)

− 0.14(0.20)

 Aging society area β2

− 0.37(0.30)

− 0.45(0.25)

− 0.53(0.25)

 General area β3

0.42(0.24)

0.40(0.21)

0.33(0.20)

 Newly developed area β4

1.34***(0.26)

0.77**(0.24)

0.66*(0.24)

 Medium-density urban area β5

1.94***(0.32)

1.31***(0.29)

1.21***(0.29)

 High-density urban area β6

2.47***(0.33)

1.44***(0.33)

1.36***(0.32)

Performance of model fitting

 AIC

259.25

239.52

237.17

 R-square

0.708

  1. The value in parentheses is standard error
  2. *p value < 0.05; **p value < 0.01; ***p value < 0.001
  3. a“Remote area” as reference category