Table 1 Average parameter estimates in the DDE model for different values of \(\tau\), the number of data points used. Results are shown for the initial number of immune individuals \(R_{t=0}\), the initial number of exposed individuals \(E_{t=0}\), the infectivity coefficient \(\beta _I\), the probability \(\rho _{noE}\) that infected individuals bypass the E stage, the residence times \(T_E\) and \(T_I\) in the exposed and infectious stages, respectively, and the average progression rates \(\delta\) (from B to C) and \(\epsilon\) (from C to R). The average estimates for the basic reproduction number \(R_0\) and the total attack rate AR are also included. Averages are based on \(M=100\) simulations (out of \(10^5\)) demonstrating the best fit on the first \(\tau\) time steps.

From: Model fit vs. predictive reliability: a case study of the 1978 influenza outbreak

 

\(R_{t=0}\)

\(E_{t=0}\)

\(\beta _I\)

\(\rho _{noE}\)

\(T_E\)

\(T_I\)

\(\delta\)

\(\epsilon\)

\(R_0\)

AR

\(\tau =14\)

248

0.03

4.38

0.68

2.98

1.72

0.33

0.52

7.51

512

\(\tau =9\)

251

1.17

4.22

0.68

3.11

3.44

0.33

0.55

14.53

512

\(\tau =6\)

427

0.87

22.56

0.00

1.20

0.85

0.27

0.85

19.09

336

\(\tau =5\)

57

0.23

15.44

0.03

0.81

0.32

0.29

0.25

3.86

647