MRTS {stratification}

Simulated Data from the Monthly Retail Trade Survey (MRTS) of Statistics Canada

Description

This data set is a vector containing the simulated values of a realistic stratification variable: the size measure used for Canadian retailers in the Monthly Retail Trade Survey (MRTS) carried out by Statistics Canada. This size measure is created using a combination of independent survey data and three administrative variables from the corporation tax return. The MRTS aims at estimating sales from retailers, which are a key monthly indicator of consumer purchasing patterns in Canada.

Usage

MRTS

Format

The format is: num [1:2000] 141 209 238 257 261 ...

Source

The data set has been simulated with the command:
exp(rst(n=2000, location=9.7093354, scale=0.7885551,
shape=-0.6384867, df=5.6243544)).
The function rst comes from the package sn. It generates random numbers for the skew-t distribution. The parameters of the distribution have been estimated with the assistance of Michel Ferland from Statistics Canada to be representative of the measure of size used in the MRTS.

References

Baillargeon, S., Rivest, L.-P., Ferland, M. (2007). Stratification en enquetes entreprises : Une revue et quelques avancees. Proceedings of the Survey Methods Section, 2007 SSC Annual Meeting.

Examples

# Production of results similar to those in Table 1 of Baillargeon, Rivest
# and Ferland (2007). The results are not the same because calculations in
# the paper were conducted on real data whereas, for confidentiality reason, 
# the MRTS data included in the package is simulated.
geo <- strata.geo(x=MRTS, CV=0.01, Ls=4, alloc=c(0.5,0,0.5))
geo
aRRMSE.geo <- var.strata(geo, model="loglinear",
              model.control=list(beta=0.9, sig2=0.015, ph=c(0.8,0.9,0.95,1)))
aRRMSE.geo$RRMSE
plot(geo, logscale=TRUE)
# The geometric method does not perform well because of some small units

cumrootf <- strata.cumrootf(x=MRTS, nclass=500, CV=0.01, Ls=4, alloc=c(0.5,0,0.5))
cumrootf
aRRMSE.cum <- var.strata(cumrootf, rh=c(0.85,0.9,0.9,1), model="loglinear",
              model.control=list(beta=0.9, sig2=0.015, ph=c(0.8,0.9,0.95,1)))
aRRMSE.cum$RRMSE

LH <- strata.LH(x=MRTS, CV=0.01, Ls=4, alloc=c(0.5,0,0.5), takeall=1, algo="Sethi")
LH
aRRMSE.LH <- var.strata(LH, rh=c(0.85,0.9,0.9,1), model="loglinear",
             model.control=list(beta=0.9, sig2=0.015, ph=c(0.8,0.9,0.95,1)))
aRRMSE.LH$RRMSE

LH.full <- strata.LH(x=MRTS, CV=0.01, Ls=4, alloc=c(0.5,0,0.5), takeall=1,
           algo="Sethi", rh=c(0.85,0.9,0.9,1), model="loglinear",
           model.control=list(beta=0.9, sig2=0.015, ph=c(0.8,0.9,0.95,1)))
LH.full
aRRMSE.LH.full <- var.strata(LH.full, rh=c(0.85,0.9,0.9,1), model="loglinear",
                  model.control=list(beta=0.9, sig2=0.015, ph=c(0.8,0.9,0.95,1)))
aRRMSE.LH.full$RRMSE