Calibrates and Reweights Units in Samples

Provides user-friendly tools for calibration in survey sampling. The package is production-oriented, and its interface is inspired by the famous popular macro 'Calmar' for SAS, so that 'Calmar' users can quickly get used to 'icarus'. In addition to calibration (with linear, raking and logit methods), 'icarus' features functions for calibration on tight bounds and penalized calibration.

Icarus (Icarus Calibrates And Reweights Units in Samples) is an R package providing useful functions for calibration and reweighting estimators in survey sampling. The former name of this package was gaston.


To cite Icarus in publications use: Rebecq, Antoine (2017). Icarus: an R package for calibration in survey sampling. R package version 0.3.0.


You can use the following instruction to install icarus (from CRAN):


However, if you wish to install the latest version of icarus, you can use devtools and install directly from this github repo:


Short example

In this example, we perform calibration (with the "raking" method) on the test dataset data_ex2 included in icarus:


N <- 300 ## Population size
## Compute the Horvitz-Thompson estimator (returns 1.666667)
weightedMean(data_ex2$cinema, data_ex2$poids, N)

## Add calibration margins
mar1 <- c("categ",3,80,90,60)
mar2 <- c("sexe",2,140,90,0)
mar3 <- c("service",2,100,130,0)
mar4 <- c("salaire", 0, 470000,0,0)
margins <- rbind(mar1, mar2, mar3, mar4)
## Compute calibration weights
wCal <- calibration(data=data_ex2, marginMatrix=margins, colWeights="poids"
                           , method="raking", description=FALSE)
## Value of the calibrated estimator: 2.471917
weightedMean(data_ex2$cinema, wCal, N)


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.3.1 by Antoine Rebecq, a year ago

Browse source code at

Authors: Antoine Rebecq [aut, cre]

Documentation:   PDF Manual  

Task views: Official Statistics & Survey Methodology

GPL-3 license

Suggests testthat, ggplot2, Rglpk, slam, xtable, magrittr

See at CRAN