Skip to main content

Table 2 Results of zero-truncated Poisson regression modelling for the offset-model (Model 1: offset is 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) for the years 2002, 2003, and 2004 based upon the SND data. is the scrapie-affected population estimated by means of the generalized Horvitz-Thompson estimator and n the number of holdings confirmed with scrapie by the SND in each year.

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

 

α

(S.E., 1Z)

β 1

β 2

log-likelihood (AIC, BIC)

2002

   

(n = 144)

 

Model 1

-4.3160 (0.0554, -77.92)

1 (fixed)

-

327

-686.43 (1374.9, 1377.7)

Model 2

0.8721 (0.1857, 4.70)

0.0560 (0.0386, 3.43)

-

151

-431.81 (867.6, 873.3)

Model 3

-0.9608 (0.6344, -1.51)

0.9498 (0.2768, 3.43)

-0.1002 (0.0298, -3.36)

153

-423.10 (852.2, 860.7)

2003

   

(n = 134)

 

Model 1

-4.3845 (0.0571, -76.78)

1

-

359

-700.69 (1544.3, 1547.1)

Model 2

1.2485 (0.1889, 6.61)

-0.0239 (0.0405, -0.59)

-

139

-399.90 (803.8, 809.4)

Model 3

-0.8759 (0.6186, -1.42)

1.0482 (0.2841, 3.69)

-0.1252 (0.0324, -3.86)

147

-388.96 (783.9, 792.3)

2004

   

(n = 151)

 

Model 1

-4.5671 (0.0653, -69.98)

1 (fixed)

-

363

-439.16 (880.3, 883.2)

Model 2

0.5373 (0.2810, 1.91)

0.0566 (0.0573, 0.99)

 

170

-316.82 (637.6, 643.4)

Model 3

-1.0624 (0.9407, -1.13)

0.7776 (0.3953, 1.97)

-0.0773 (0.0412, -1.88)

172

-314.68 (635.3, 644.1)

  1. 1Z = coeff./S.E.