calc_nqx: Calculate the probability of dying between age x and x+n...

In mrc-ide/demogsurv: Demographic analysis of DHS and other household surveys

Calculate the probability of dying between age x and x+n (nqx)

Description

Default arguments are configured to calculate under 5 mortality from a DHS Births Recode file.

Usage

calc_nqx(
  data,
  by = NULL,
  agegr = c(0, 1, 3, 5, 12, 24, 36, 48, 60)/12,
  period = NULL,
  cohort = NULL,
  tips = c(0, 5, 10, 15),
  clusters = ~v021,
  strata = ~v024 + v025,
  weight = "v005",
  dob = "b3",
  dod = "dod",
  death = "death",
  intv = "v008",
  varmethod = "lin",
  origin = 1900,
  scale = 12
)

Arguments

data	A dataset (data.frame), for example a DHS births recode (BR) dataset.
by	A formula specifying factor variables by which to stratify analysis.
agegr	Numeric vector defining ages in years for splits.
period	Numeric vector defining calendar periods to stratify analysis, use NULL for no periods.
cohort	Numeric vector defining birth cohorts to stratify analysis, use NULL for no cohort stratification.
tips	Break points for TIme Preceding Survey.
clusters	Formula or data frame specifying cluster ids from largest level to smallest level, ‘~0’ or ‘~1’ is a formula for no clusters.
strata	Formula or vector specifying strata, use ‘NULL’ for no strata.
weight	Formula or vector specifying sampling weights.
dob	Variable name for date of birth (character string).
dod	Variable name for date of death (character string).
death	Variable name for event variable (character string).
intv	Variable name for interview date (character string).
varmethod	Method for variance calculation. Currently "lin" for Taylor linearisation or "jk1" for unstratified jackknife, or "jkn", for stratified jackknife.
origin	Origin year for date arguments. 1900 for CMC inputs.
scale	Scale for dates inputs to calendar years. 12 for CMC inputs.

Examples

data(zzbr)
zzbr$death <- zzbr$b5 == "no"  # b5: child still alive ("yes"/"no")
zzbr$dod <- zzbr$b3 + zzbr$b7 + 0.5

## Calculate 5q0 from birth history dataset.
## Note this does NOT exactly match DHS calculation.
## See calc_dhs_u5mr().
u5mr <- calc_nqx(zzbr)
u5mr

## Retrieve sample covariance and correlation
vcov(u5mr)  # sample covariance
cov2cor(vcov(u5mr))  # sample correlation

## 5q0 by sociodemographic characteristics
calc_nqx(zzbr, by=~v102) # by urban/rural residence
calc_nqx(zzbr, by=~v190, tips=c(0, 10)) # by wealth quintile, 0-9 years before
calc_nqx(zzbr, by=~v101+v102, tips=c(0, 10)) # by region and residence

## Compare unstratified standard error estiamtes for linearization and jackknife
calc_nqx(zzbr, varmethod = "lin")  # unstratified design
calc_nqx(zzbr, strata=NULL, varmethod = "lin")  # unstratified design
calc_nqx(zzbr, strata=NULL, varmethod = "jk1")  # unstratififed jackknife
calc_nqx(zzbr, varmethod = "jkn")  # stratififed jackknife

## Calculate various child mortality indicators (neonatal, infant, etc.)
calc_nqx(zzbr, agegr=c(0, 1)/12)  # neonatal
calc_nqx(zzbr, agegr=c(1, 3, 5, 12)/12) # postneonatal
calc_nqx(zzbr, agegr=c(0, 1, 3, 5, 12)/12) # infant (1q0)
calc_nqx(zzbr, agegr=c(12, 24, 36, 48, 60)/12) # child (4q1)
calc_nqx(zzbr, agegr=c(0, 1, 3, 5, 12, 24, 36, 48, 60)/12) # u5mr (5q0)

## Calculate annaul 5q0 by calendar year
calc_nqx(zzbr, period=2005:2015, tips=NULL)

mrc-ide/demogsurv documentation built on March 21, 2022, 9:49 p.m.