est.prev: Estimated overall prevalence from sample selection models
Description
est.prev can be used to calculate the overall estimated prevalence from a sample selection model, with corresponding confidence intervals
obtained using the delta method.
A fitted SemiParBIVProbit object as produced by SemiParBIVProbit().
sig.lev
Significance level.
sw
Survey weights.
naive
It indicates whether the prevalence is calculated using the (naive/classic imputation) probit or sample selection model. This option has been introduced to compared adjusted and unadjusted estimates.
Value
resIt returns three values: lower confidence interval limit, estimated prevalence and upper confidence interval limit.
sig.levSignificance level used.
Details
est.prev estimates the overall prevalence of a disease (e.g. HIV) when there are missing values that are not at random.