This function allows to build the information regarding strata in the population required as an input by the algorithm of Bethel for the optimal allocation. In order to estimate means and standard deviations for target variables Y's, we need data coming from: (1) a previous round of the survey whose sample we want to plan; (2) sample data from a survey with variables that are proxy to the ones we are interested to; (3) a frame containing values of Y's variables (or proxy variables) for all the population. In all cases, each unit in the dataset must contain auxiliary information (X's variables) and also target variables Y's (or proxy variables) values: under these conditions it is possible to build the dataframe "strata", containing information on the distribution of Y's in the different strata (namely, means and standard deviations), together with information on strata (total population, if it is to be censused or not, the cost per single interview). If the information is contained in a sample dataset, a variable named WEIGHT is expected to be present. In case of a frame, no such variable is given, and the function will define a WEIGHT variable for each unit, whose value is always '1'. Missing values for each Y variable will not be taken into account in the computation of means and standard deviations (in any case, NA's can be present in the dataset). The dataframe "strata" is written to an external file (tab delimited, extension "txt"), and will be used as an input by the function "optimizeStrata".

buildStrataDF(dataset,
model=NULL,
progress=TRUE,
verbose=TRUE)

## Arguments

dataset 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 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. If set to TRUE, a progress bar is visualised during the execution. Default is TRUE. If set to TRUE, information is given about the number of strata generated. Default is TRUE.

## Value

A dataframe containing strata

Giulio Barcaroli

## Examples

if (FALSE) {
# Plain example without model
data(swissframe)

# More complex example with models
library(SamplingStrata)
data(swissmunicipalities)
swiss <- swissmunicipalities[,c("HApoly","Surfacesbois","Airind","POPTOT")]
Y1 = swiss$Surfacesbois X1 = swiss$HApoly
mod1 <- lm( Y1 ~ X1 )
summary(mod1)
mod1$coefficients[2] summary(mod1)$sigma

Y2 = swiss$Airind X2 = swiss$POPTOT
plot(log(X2[X2>0]),log(Y2[X2>0]))
mod2 <- lm( log(Y2[X2 > 0 & Y2>0]) ~ log(X2[X2 > 0 & Y2>0]) )
summary(mod2)
mod2$coefficients[2] summary(mod2)$sigma

swiss$id <- c(1:nrow(swiss)) swiss$dom <- 1
frame <- buildFrameDF(swiss,id="id",X="id",Y=c("HApoly","POPTOT"),domainvalue="dom")

model <- NULL
model$type[1] <- "linear" model$beta[1] <- mod1$coefficients[2] model$sig2[1] <- summary(mod1)$sigma model$gamma[1] = 2
model$type[2] <- "loglinear" model$beta[2] <- mod2$coefficients[2] model$sig2[2] <- summary(mod2)$sigma model$gamma[2] = NA
model <- as.data.frame(model)