svyPVpack (version 0.1-1)

svyPVglm: Survey-weighted generalised linear models with uasge of palusible values.

Description

Fit a generalised linear model to data from a complex survey design with usage of plausible values.

Usage

svyPVglm(formula, design, placeholder = 1:10, family = gaussian())
"summary"(object, ...)

Arguments

formula

a model formula (all variables have to be part of a survey design objecct created by the survey package). For a notation description for the plausible values see in 'details'.

design

a survey design which was generated by the survey package

placeholder

a vector of symbols, which were used for numbering the plausible values. For a detailed description see in 'details'.

family

a description of the error distribution and link function to be used in the model.This has to be either a character string naming a family function, a family function or the result of a call to a family function.

object

A object of class svyPVglm.

...

Value

coef

Shows the regression coefficients and test statistics.

mod.fit

Shows the model test statistic. For more details see svyglm.

Details

All variables mentioned in the formula object must be part of the survey design object. Instead of the symbols, which were used for numbering the plausible values use '..' as notation (e.g. placeholder = 1:5 and PVLIT.. stands for PVLIT1, PVLIT2, PVLIT3, PVLIT4, PVLIT5). Missing values are deleted listwise.

References

Lumley, T. (2010). Complex Surveys. Hoboken, NJ: Wiley.

Saerndal, C.-E. & Swensson, B. & Wretman, J. (1992). Model Assisted Survey Sampling. New York: Springer.

Chaudhuri, A. & Stenger, H. (2005). Survey Sampling. Theory and Methods. Boka Raton, FL: Chapman & Hall/CRC.

See Also

svyglm, svyPVcor, svyPVeta

Examples

data(svy_example1)

res_glm <- svyPVglm(plaus.. ~ otherPlaus.. + age, svy.exrep, placeholder = 1:3)

res_glm