Skip to main content

Table 1 Relative cross validation error for single hold-out strata in all-cause, pneumonia- and injury-specific mortality by age in China, 1996–2015

From: Adapting and validating the log quadratic model to derive under-five age- and cause-specific mortality (U5ACSM): a preliminary analysis

Age Relative error in \(_{x}q_{0}\) per 1000 livebirths  xq0 (range)
Log quadratic with \(k =0\) (%) Log quadratic with estimated k (%) Standard\(^{\ddagger }\) (%)
All cause\(^{\dagger }\)
 0–6 days (\(_{6d}q_{0}\)) 18 16 67 1.3–29.7
 0–27 days (\(_{27d}q_{0}\)) 15 0 0 1.9–38.0
 0–5 months (\(_{5mo}q_{0}\)) 7 3 17 2.6–55.3
 0–11 months (\(_{11mo}q_{0}\)) 5 4 18 2.9–62.4
 0–23 months (\(_{23mo}q_{0}\)) 3 3 16 3.2–68.4
 0–59 months (\(_{59mo}q_{0}\)) 0 0 0 3.6–76.7
Pneumonia-specific*
 0–6 days (\(_{6d}q_{0,c}\)) 78 64 51 0.0–3.5
 0–27 days (\(_{27d}q_{0,c}\)) 47 0 0 0.0–6.9
 0–5 months (\(_{5mo}q_{0,c}\)) 14 11 36 0.1–16.0
 0–11 months (\(_{11mo}q_{0,c}\)) 9 8 37 0.1–20.0
 0–23 months (\(_{23mo}q_{0,c}\)) 5 5 30 0.2–22.9
 0–59 months (\(_{59mo}q_{0,c}\)) 0 0 0 0.2–24.0
Injury-specific**
 0–6 days (\(_{6d}q_{0,c}\)) 68 62 59 0.0–4.4
 0–27 days (\(_{27d}q_{0,c}\)) 68 0 0 0.0–4.7
 0–5 months (\(_{5mo}q_{0,c}\)) 43 25 31 0.0–6.1
 0–11 months (\(_{11mo}q_{0,c}\)) 32 23 23 0.0–6.5
 0–23 months (\(_{23mo}q_{0,c}\)) 16 17 18 0.1–7.4
 0–59 months (\(_{59mo}q_{0,c}\)) 0 0 0 0.3–11.9
  1. Error shown as average percent difference between estimated and observed \(_{x}q_{0}\) and \(_{x}q_{0,c}\)
  2. \({\ddagger }\)Based on constant mortality daily/monthly rate across age within 0–27 days and 1–59 months
  3. \(\dagger\) \(k\) matched to all-cause neonatal mortality
  4. *k matched to pneumonia-specific neonatal mortality
  5. **k matched to injury-specific neonatal mortality