Builds the "errors" dataframe containing information on the precision constraints set on the target variables

This function allows to build the dataframe containing the precision constraints to be set on the target variables of the sampling survey. It can be applied to the frame dataframe or to the strata dataframe, once they have been defined. The dataframe values of the precision constraints will not be initialized, so the user has to do it.

buildErrorsDF(frame=NULL,
              strata=NULL)

Arguments

frame	This is the name of the dataframe containing the sample data, or the frame data. It is strictly required that auxiliary information is organised in variables named as X1, X2, ... , Xm (there should be at least one of them) and the target variables are denoted by Y1, Y2, ... , Yn. In addition, in case of sample data, a variable named 'WEIGHT' must be present in the dataframe, containing the weigths associated to each sampling unit
strata	In case the Y variables are not directly observed, but are estimated by means of other explicative variables, in order to compute the anticipated variance, information on models are given by a dataframe "model" with as many rows as the target variables. Each row contains the indication if the model is linear o loglinear, and the values of the model parameters beta, sig2, gamma (> 1 in case of heteroscedasticity). Default is NULL.

frame

This is the name of the dataframe containing the sample data, or the frame data. It is strictly required that auxiliary information is organised in variables named as X1, X2, ... , Xm (there should be at least one of them) and the target variables are denoted by Y1, Y2, ... , Yn. In addition, in case of sample data, a variable named 'WEIGHT' must be present in the dataframe, containing the weigths associated to each sampling unit

strata

In case the Y variables are not directly observed, but are estimated by means of other explicative variables, in order to compute the anticipated variance, information on models are given by a dataframe "model" with as many rows as the target variables. Each row contains the indication if the model is linear o loglinear, and the values of the model parameters beta, sig2, gamma (> 1 in case of heteroscedasticity). Default is NULL.

Value

A dataframe containing precision constraints on the target variables.

Author

Giulio Barcaroli

Examples

if (FALSE) {
library(SamplingStrata)
data(swissmunicipalities)
swissmunicipalities$HApoly.cat <- var.bin(swissmunicipalities$HApoly,15)
swissmunicipalities$POPTOT.cat <- var.bin(swissmunicipalities$POPTOT,15)
frame <- buildFrameDF(df = swissmunicipalities,
                       id = "COM",
                       X = c("POPTOT.cat","HApoly.cat"),
                       Y = c("Airbat","Surfacesbois"),
                       domainvalue = "REG")
strata <- buildStrataDF(frame)
ndom <- length(unique(swissmunicipalities$REG))
cv <- buildErrorsDF(strata=strata)
#cv <- buildErrorsDF(frame=frame)
cv$CV1 <- c(rep(0.03,7))
cv$CV2 <- c(rep(0.05,7))
}