Sensitivity of contact-tracing for COVID-19 in Thailand: a capture-recapture application

BMC Infectious Diseases

Table 5 Model performance for the count index cases with infected contacts

Model	Log-likelihood	AIC	BIC
Poisson (\({\hat{\lambda }}\) = 1.126)	\(-\)32.6684	67.3368	68.738
Negative Binomial (mue = 1.176, size = 1175.709)	\(-\)32.6914	69.3829	72.1853
Geometric (\({\hat{p}}\) = 0.6)	\(-\)33.6506	69.3012	70.7024

Bold values indicate show that Poisson is the best fit for the distribution of the number of index cases with infected contacts (n=30) due to the smallest AIC and BIC

ISSN: 1471-2334