Confidence intervals for regression parameters

Computes confidence intervals for regression parameters in svyglm objects. The default is a Wald-type confidence interval, adding and subtracting a multiple of the standard error. The method="likelihood" is an interval based on inverting the Rao-Scott likelihood ratio test. That is, it is an interval where the working model deviance is lower than the threshold for the Rao-Scott test at the specified level.

Usage

# S3 method for class 'svyglm'
confint(object, parm, level = 0.95, method = c("Wald", "likelihood"), ddf = NULL, ...)

Arguments

object: svyglm object
parm: numeric or character vector indicating which parameters to construct intervals for.
level: desired coverage
method: See description above
ddf: Denominator degrees of freedom for "likelihood" method, to use a t distribution rather than normal. If NULL, use object$df.residual for Taylor-series standard errors, or object$df.coef for Bell-McCaffrey standard errors with adjusted degrees of freedom.
...: for future expansion

Value

A matrix of confidence intervals, possibly with additional attributes levels and/or degf.

References

J. N. K. Rao and Alastair J. Scott (1984). On Chi-squared Tests For Multiway Contigency Tables with Proportions Estimated From Survey Data. Annals of Statistics 12:46-60.

Robert M. Bell and Daniel F. McCaffrey (2002). Bias Reduction in Standard Errors for Linear Regression with Multi-Stage Samples. Survey Methodology 28 (2), 169-181. https://www150.statcan.gc.ca/n1/pub/12-001-x/2002002/article/9058-eng.pdf

Examples

data(api)
dclus2<-svydesign(id=~dnum+snum, fpc=~fpc1+fpc2, data=apiclus2)

m<-svyglm(I(comp.imp=="Yes")~stype*emer+ell, design=dclus2, family=quasibinomial)
confint(m)
#>                   2.5 %     97.5 %
#> (Intercept)  0.56476264 3.70319468
#> stypeH      -4.28558148 0.93393621
#> stypeM      -3.70649097 1.34532080
#> emer        -0.08588860 0.04659558
#> ell         -0.05693166 0.02448049
#> stypeH:emer -0.13853776 0.07244824
#> stypeM:emer -0.20697834 0.14040185
confint(m, method="like",ddf=NULL, parm=c("ell","emer"))
#>            2.5 %     97.5 %
#> ell  -0.05891077 0.02469912
#> emer -0.08637239 0.04994329
#> attr(,"levels")
#> [1] 0.95

m2<-svyglm(I(comp.imp=="Yes")~stype*emer+ell, design=dclus2, family=quasibinomial,
    std.errors="Bell-McCaffrey-2", degf=TRUE)
confint(m2)
#>                   2.5 %     97.5 %
#> (Intercept)  0.08312364 4.18483368
#> stypeH      -4.96618469 1.61453942
#> stypeM      -4.38389918 2.02272901
#> emer        -0.12986603 0.09057301
#> ell         -0.11673141 0.08428024
#> stypeH:emer -0.17404101 0.10795149
#> stypeM:emer -0.24275741 0.17618092
#> attr(,"degf")
#> (Intercept)      stypeH      stypeM        emer         ell stypeH:emer 
#>    4.588330    5.490844    5.104007    4.018683    1.949084    5.630433 
#> stypeM:emer 
#>    6.771212