Skip to main content

Table 1 Distributional assumptions regarding the underlying age-specific malaria FOI

From: Modelling longitudinal binary outcomes with outcome dependent observation times: an application to a malaria cohort study

Distribution

\(\varvec{\theta }\)

\(h(a_{ij}; \varvec{\theta })\)

\(\lambda _0(a_{ij})\)

Exponential

\(\theta _1 > 0\)

\(\log (\theta _1 a_{ij})\)

\(\theta _1\)

Weibull

\(\theta _1, \theta _2 > 0\)

\(\log (\theta _1 a_{ij}^{\theta _2})\)

\(\theta _1 \theta _2 a_{ij}^{\theta _2-1}\)

Gompertz

\(\theta _1 > 0, -\infty< \theta _2 < +\infty\)

\(\log \left[ \frac{\theta _1}{\theta _2} \left( e^{\theta _2 a_{ij}}-1\right) \right]\)

\(\theta _1 e^{\theta _2 a_{ij}}\)

Log-logistic

\(\theta _1, \theta _2 > 0\)

\(\log \left\{ \log \left[ 1+\left( \theta _1 a_{ij}\right) ^{\theta _2}\right] \right\}\)

\(\frac{\theta _1 \theta _2 (\theta _1 a_{ij})^{\theta _2 - 1}}{1 + (\theta _1 a_{ij})^{\theta _2}}\)

Fractional polynomial

\(\theta _2 < 0\)

\(\theta _2 a_{ij}^{-1}\)

\(-\theta _2 a_{ij}^{-2} e^{\theta _2 a_{ij}^{-1}}\)