weights.Rd
Extracts the current weights of units belonging to a survey design object.
weights(object, ...)
object | Object of class |
---|---|
... | Arguments for future expansion. |
The current weights of object
are, by definition, those weights that would be used for estimation purposes on that object (e.g. by functions svystatTM
, svystatR
, svystatS
, svystatSR
, svystatQ
, svystatB
, svystatL
, ...). The nature of such weights depends on the class of object
: calibrated weights for class cal.analytic
and direct weights otherwise.
A vector of weights, whose components are positionally tied to the sampling units belonging to object
.
If object
has undergone multiple, subsequent calibration steps, the function will return the output weights generated by the last calibration step.
Function g.range
to asses the range of the g-weights of a calibrated design object.
# Creation of the object to be calibrated: data(data.examples) exdes<-e.svydesign(data=example,ids=~towcod+famcod,strata=~SUPERSTRATUM, weights=~weight) # Retrieve the weights and summarize their distribution: summary(weights(exdes))#> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 95.5 218.8 321.3 308.0 362.3 931.2# Now calibrate (global solution) on the joint distribution of sex # and marstat (totals in pop03): excal.1st<-e.calibrate(design=exdes,df.population=pop03, calmodel=~marstat:sex-1,calfun="linear",bounds=bounds) # Retrieve the current weights (i.e. the calibrated ones) and # summarize their distribution: summary(weights(excal.1st))#> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 94.44 218.49 317.75 306.10 361.09 932.01# Now calibrate once again, this time on the marginal distribution # of age in 5 classes (age5c) inside provinces (procod) (totals in pop06p) # with the partitioned solution, the logit distance and bounds=c(0.5, 1.5): excal.2nd<-e.calibrate(design=excal.1st,df.population=pop06p, calmodel=~age5c-1,partition=~procod,calfun="logit", bounds=c(0.5, 1.5)) # Notice that the print method correctly takes the calibration chain # into account: excal.2nd#> Calibrated, Stratified 2 - Stage Cluster Sampling Design (with replacement) #> - [55] strata #> - [1307, 2372] clusters #> #> Call: #> 2: e.calibrate(design = excal.1st, df.population = pop06p, calmodel = ~age5c - #> 1, partition = ~procod, calfun = "logit", bounds = c(0.5, #> 1.5)) #> 1: e.calibrate(design = exdes, df.population = pop03, calmodel = ~marstat:sex - #> 1, calfun = "linear", bounds = bounds)# Now retrieve the current weights (i.e. the ones generated by the second # calibration step) and summarize their distribution: summary(weights(excal.2nd))#> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 95.98 231.59 294.78 312.56 370.24 880.11