msvyratio: Robust ratio M-estimation for complex samples

In rhte: Robust Horvitz-Thompson Estimation

Description

msvyratio computes a robust ratio estimate for complex samples by means of M-estimation ("rht" type).

Usage

msvyratio(numerator, denominator, design, k, na.rm = FALSE, 
          control = rht.control(...), ...)

Arguments

numerator	a formula object (only one variable)
denominator	a formula object (only one variable)
design	a survey.design object
k	robustness tuning constant
na.rm	should cases with missing values be dropped? (default FALSE)
control	control object; see rht.control
...	(additional specifications which are delivered to rht.control)

Details

msvyratio computes a robust ratio estimate for complex samples by means of M-estimation (type "rht"; see msvymean for more details). Variance estimates are computed as first-order linearization using the design-based estimation facilities in the survey package.

The initial value is a weighted median or a ratio of weighted medians. You may set steps equal to one in order to obtain an one-step estimator.

msvyratio allows also the estimation for domains. Use the command subset and a design subset expression instead of the original survey.design object in msvyratio (see examples for more details).

Value

Object of class "svystat.rob"

The following (S3) methods are defined for objects of the class "svystat.rob":

• print method,

• summary method,

• coef method,

• vcov method,

• residuals method,

• robweights method.

Author(s)

Beat Hulliger and Tobias Schoch

References

Hulliger, B. (1995): Outlier robust Horvitz-Thompson estimators, Survey Methodology 21 (1), pp. 79-87.

Hulliger, B. (1999): Simple and robust estimators for sampling, Proceedings of the Survey Research Methods Section, American Statistical Association, 1999, pp. 54-63.

Hulliger, B. and T. Schoch (2011): Elementary robust estimators. In: Robust methodology for Laeken indicators: AMELI Deliverable D4.2, ed. by B. Hulliger, A. Alfons, P. Filzmoser, A. Meraner, T. Schoch and M. Templ. AMELI Project.

See Also

svyratio

Examples

## load "api" data set from "survey" package (a description of the data
## set can be found there)
data(api)
## define "survey.design" for stratified sampling
dstrat <- svydesign(id=~1,strata=~stype, weights=~pw, data=apistrat, 
fpc=~fpc)
## compute a robust Horvitz-Thompson estimate for the mean of the 
## variable api00 (Academic Performance Index in 2000)
ratio1 <- msvyratio(~api00, ~api99, dstrat, k=1.2, na.rm=TRUE)
## get a summary of the estimation
summary(ratio1)

rhte documentation built on May 2, 2019, 6:24 p.m.