Table 1 Overview of the models used to create predictions
From: Comparing methods to predict baseline mortality for excess mortality calculations
Name | Model |
|---|---|
Average | \({Y}_{t} \sim NegBin\left({\mu }_{t},\theta \right)\) \(\mathrm{log}\left({\mu }_{t}\right)= {\beta }_{0}+{f}_{cc}\left(w\left[t\right]\right)\) |
Linear | \({Y}_{t} \sim NegBin\left({\mu }_{t},\theta \right)\) \(\mathrm{log}\left({\mu }_{t}\right)= {\beta }_{0}+{\beta }_{1}t+{f}_{cc}\left(w\left[t\right]\right)\) |
WHO | \({Y}_{t} \sim NegBin\left({\mu }_{t},\theta \right)\) \(\mathrm{log}\left({\mu }_{t}\right)= {f}_{tp}\left(t\right)+{f}_{cc}\left(w\left[t\right]\right)\) |
Acosta–Irizarry | \({Y}_{t} \sim QuasiPoi\left({\mu }_{t}\right)\) \(\mathrm{log}\left({\mu }_{t}\right)=\left\{\begin{array}{c} {\beta }_{0}+{f}_{nc}\left(t\right)+{\sum }_{k=1}^{2}\left[\mathrm{sin}\left(2\pi \cdot k\cdot w\left[t\right]\right)+\mathrm{cos}\left(2\pi \cdot k\cdot w\left[t\right]\right)\right]*\\ {\beta }_{0}+{\beta }_{1}t+{\sum }_{k=1}^{2}\left[\mathrm{sin}\left(2\pi \cdot k\cdot w\left[t\right]\right)+\mathrm{cos}\left(2\pi \cdot k\cdot w\left[t\right]\right)\right]**\end{array}\right.\) |