Skip to main content

Table 4 Results of zero-truncated Poisson regression modelling including population size estimates (columns 4 contains the robust Zelterman estimate in brackets) for the offset-model (Model 1: offset if log-size of holding), for the model treating log-size as a free covariate (Model 2) and the model with a quadratic term included (Model 3) based upon the CSFS data (n = 214).

From: On the question of proportionality of the count of observed Scrapie cases and the size of holding

Model 1

coefficient

S.E./Z = coeff./S.E.

(, 95% CI)

log-likelihood (AIC, BIC)

α

-6.5573

0.0785/-83.48

457 (1077, 0 – 12,882)

-274.38

β 1

1 (fixed)

  

(550.76, 554.13)

Model 2

coefficient

S.E./Z = coeff./S.E.

(, 95% CI)

log-likelihood (AIC, BIC)

α

-3.8869

0.6123/-6.35

197 (398, 199 – 597)

-265.92

β 1

0.6164

0.0893/6.90

 

(535.83, 542.57)

Model 3

coefficient

S.E./Z = coeff./S.E.

(, 95% CI)

log-likelihood (AIC, BIC)

α

1.6045

1.0116/1.59

228 (403, 200 – 606)

-257.76

β 1

-1.2240

0.3430/-3.57

 

(521.52, 531.61)

β 2

0.1486

0.0291/5.11