regTermTest {survey}  R Documentation 
Provides Wald test and working Wald and working likelihood ratio (RaoScott) test of the
hypothesis that all coefficients associated with a particular
regression term are zero (or have some other specified
values). Particularly useful as a substitute for anova
when not fitting by maximum likelihood.
regTermTest(model, test.terms, null=NULL,df=NULL, method=c("Wald","WorkingWald","LRT"), lrt.approximation="saddlepoint")
model 

test.terms 
Character string or onesided formula giving name of term or terms to test 
null 
Null hypothesis values for parameters. Default is zeros 
df 
Denominator degrees of freedom for an F test. If

method 
If 
lrt.approximation 
method for approximating the distribution of
the LRT and Working Wald statistic; see 
The Wald test uses a chisquared or F distribution. The two
workingmodel tests come from the (misspecified) working model where the
observations are independent and the weights are frequency weights. For
categorical data, this is just the model fitted to the estimated
population crosstabulation. The RaoScott LRT statistic is the likelihood
ratio statistic in this model. The working Wald test statistic is the Wald statistic
in this model. The workingmodel tests do not have a chisquared
sampling distribution: we use a linear combination of chisquared or F
distributions as in pchisqsum
. I believe the working Wald
test is what SUDAAN refers to as a
"Satterthwaite adjusted Wald test".
To match other software you will typically need to use lrt.approximation="satterthwaite"
An object of class regTermTest
or regTermTestLRT
.
The "LRT"
method will not work if the model had starting values supplied for the regression coefficients. Instead, fit the two models separately and use anova(model1, model2, force=TRUE)
Rao, JNK, Scott, AJ (1984) "On Chisquared Tests For Multiway Contingency Tables with Proportions Estimated From Survey Data" Annals of Statistics 12:4660.
Lumley T, Scott A (2012) "Partial likelihood ratio tests for the Cox model under complex sampling" Statistics in Medicine 17 JUL 2012. DOI: 10.1002/sim.5492
Lumley T, Scott A (2014) "Tests for Regression Models Fitted to Survey Data" Australian and New Zealand Journal of Statistics 56:114 DOI: 10.1111/anzs.12065
anova
, vcov
, contrasts
,pchisqsum
data(esoph) model1 < glm(cbind(ncases, ncontrols) ~ agegp + tobgp * alcgp, data = esoph, family = binomial()) anova(model1) regTermTest(model1,"tobgp") regTermTest(model1,"tobgp:alcgp") regTermTest(model1, ~alcgp+tobgp:alcgp) data(api) dclus2<svydesign(id=~dnum+snum, weights=~pw, data=apiclus2) model2<svyglm(I(sch.wide=="Yes")~ell+meals+mobility, design=dclus2, family=quasibinomial()) regTermTest(model2, ~ell) regTermTest(model2, ~ell,df=NULL) regTermTest(model2, ~ell, method="LRT", df=Inf) regTermTest(model2, ~ell+meals, method="LRT", df=NULL) regTermTest(model2, ~ell+meals, method="WorkingWald", df=NULL)