Horvitz-Thompson prevalence estimator

Source: R/study_estimators.R

Horvitz-Thompson prevalence estimator

get_study_est_ht(
  data,
  hidden_var = "hidden",
  weight_var = "pps_weight",
  total_var = "total",
  survey_design = ~pps_cluster + strata(pps_strata),
  n_boot = 100,
  parallel_boot = FALSE,
  prefix = "pps",
  label = "ht"
)

Arguments

data

pass-through population data frame

hidden_var

variable containing hidden group membership indicator

weight_var

variable containing sampling weights

total_var

variable containing size of population for prevalence estimation

survey_design

a formula describing the design of the survey (for bootstrap)

n_boot

number of bootstrap resamples

parallel_boot

logical, whether to compute bootstrap samples in parallel using foreach package

prefix

character prefix used for sampling variables (has to include [prefix]_weights)

label

character string describing the estimator

Value

Data frame of HT estimates for a single study

References

Salganik, Matthew J. "Variance estimation, design effects, and sample size calculations for respondent-driven sampling." Journal of Urban Health 83, no. 1 (2006): 98. https://doi.org/10.1007/s11524-006-9106-x