add.ci.est: Calculates confidence interval estimates from samples.

In david-borchers/sampling: Module MT4608 Sampling Theory Practical Package

Calculates confidence interval estimates from samples.

Description

This function calculates confidence interval estimates for population mean, total or proportion from an object of class 'sample' or 'samples.population' and them it to the sample in columns named "ci.lower" and "ci.upper".

Usage

  add.ci.est(dat,type="ybar",level=0.95)

Arguments

dat	Object of class 'sample' or 'samples.population'.
level	Confidence level; 0.95 (95%) by default.

Details

Returns an approximate confidence interval estimate of the appropriate type, assuming that the statistic has a t-distribution with degrees of freedom equal to sample size. Elements "point.est" and "var.est" are used to construct the confidence interval(s), so they must already be in 'dat'. They can be added to 'dat' using add.point.est and add.var.est

Value

The function returns an object identical to 'dat' except that elements $ci.lower and $ci.upper, being the lower and upper (100*level) confidence interval bounds from the sample, have been added.

See Also

ci.est, var.est, add.var.est, point.est, add.point.est

Examples

  data(barnett) # get barnett data
  
  # All samples total weight of class from srs:
  samp.dbn<-take.sample(barnett,y.name="weight",n=23,take.all=TRUE)
  # estimate total
  samp.dbn<-add.point.est(samp.dbn,type="ybar.tot")
  samp.dbn<-add.var.est(samp.dbn,type="ybar.tot")
  samp.dbn<-add.ci.est(samp.dbn)
  
  # proportion of class with weight more than 15, from srs:
  samp.dbn<-take.sample(barnett,y.name="weight",n=23,take.all=TRUE)
  # estimate total
  samp.dbn<-add.point.est(samp.dbn,type="ybar.ppn",cond=">15")
  samp.dbn<-add.var.est(samp.dbn,type="ybar.ppn",cond=">15")
  samp.dbn<-add.ci.est(samp.dbn)
  
  # Stratified mean
  # Define 4 strata based on weight, with 10 lb interval widths:
  strat.barnett<-define.subunit(barnett,aux.name="weight",breaks=c(0,10,20,30,40),type="strat")
  # take stratified sample
  samp.dbn<-take.sample(strat.barnett,y.name="height",type="strat",n=c(2,9,7,3),take.all=TRUE)
  # estimate mean height using stratified estimator
  samp.dbn<-add.point.est(samp.dbn,type="strat.mean")
  samp.dbn<-add.var.est(samp.dbn,type="strat.mean")
  samp.ci<-add.var.est(samp.dbn)
  
  # Cluster mean
  # Define 5 clusters based on weight:
  clust.barnett<-define.subunit(barnett,aux.name="weight",
                                breaks=c(0,10,15,20,25,30,40),type="clust")
  # take cluster sample of size 3
  samp.dbn<-take.sample(clust.barnett,y.name="height",type="clust",m=3,take.all=TRUE)
  # estimate mean height using cluster estimator
  samp.dbn<-add.point.est(samp.dbn,type="clust.mean")
  samp.dbn<-add.var.est(samp.dbn,type="clust.mean")
  samp.dbn<-add.ci.est(samp.dbn)
  
  # Ratio estimator
  # take srs of size 6, with weight as auxiliary variable:
  samp.dbn<-take.sample(barnett,y.name="height",aux.name="weight",type="srs",n=23,take.all=TRUE)
  # estimate mean height using cluster estimator
  samp.dbn<-add.point.est(samp.dbn,type="ratio.mean")
  samp.dbn<-add.var.est(samp.dbn,type="ratio.mean")
  samp.dbn<-add.ci.est(samp.dbn)
  

david-borchers/sampling documentation built on Sept. 17, 2022, 7:54 a.m.