1\name{psrsq} 2\alias{psrsq} 3%- Also NEED an '\alias' for EACH other topic documented here. 4\title{ 5Pseudo-Rsquareds 6} 7\description{ 8Compute the Nagelkerke and Cox--Snell pseudo-rsquared statistics, primarily for logistic regression. A generic function with methods for \code{glm} and \code{\link{svyglm}}. The method for \code{svyglm} objects uses the design-based estimators described by Lumley (2017) 9} 10\usage{ 11psrsq(object, method = c("Cox-Snell", "Nagelkerke"), ...) 12} 13%- maybe also 'usage' for other objects documented here. 14\arguments{ 15 \item{object}{ 16A regression model (\code{glm} or \code{svyglm}) 17} 18 \item{method}{ 19Which statistic to compute 20} 21 \item{\dots}{ 22For future expansion 23} 24} 25 26\value{ 27Numeric value 28} 29\references{ 30Lumley T (2017) "Pseudo-R2 statistics under complex sampling" Australian and New Zealand Journal of Statistics DOI: 10.1111/anzs.12187 (preprint: \url{https://arxiv.org/abs/1701.07745}) 31} 32 33\seealso{ 34\code{\link{AIC.svyglm}} 35} 36\examples{ 37data(api) 38dclus2<-svydesign(id=~dnum+snum, weights=~pw, data=apiclus2) 39 40model1<-svyglm(I(sch.wide=="Yes")~ell+meals+mobility+as.numeric(stype), 41 design=dclus2, family=quasibinomial()) 42 43psrsq(model1, type="Nagelkerke") 44 45} 46% Add one or more standard keywords, see file 'KEYWORDS' in the 47% R documentation directory. 48\keyword{survey }% use one of RShowDoc("KEYWORDS") 49\keyword{regression }% __ONLY ONE__ keyword per line 50