mmmgee (version 1.20)

mmmgee: Covariance Matrix Estimation for Multiple Marginal GEE Models

Description

Calculate the covariance matrix for a stacked vector of regression coefficients from multiple marginal GEE models fitted with geem2.

Usage

mmmgee(x, biascorr = FALSE)

Arguments

x

a list of geem objects fitted with geem2. The geem objects must be different models calculated with data from the same subjects. In particular, the parameter id in the call to geem2 must refer to the same subjects in each model.

biascorr

logical, if TRUE, a bias corrected covariance matrix is calculate by extending the method due to Mancl and DeRouen to multiple models. See references.

Value

A list with class mmmgee containing the following components:

beta

The stacked vector of regression coefficient estimates from the models in x.

V

The estimated covariance matrix of the regression coefficient estimates.

A

The outer component of \(V=ABA\).

B

The inner component of \(V=ABA\).

biascorr

The value of the input argument biascorr (logical).

n

A vector with the number of clusters in each model in x.

p

A vector with number of regression coefficients in each model in x.

References

Lloyd A. Mancl, Timothy A. DeRouen. A covariance estimator for GEE with improved small sample properties. Biometrics, 2001, 57(1):126-134.

See Also

geem2, mmmgee.test

Examples

# NOT RUN {
data(keratosis)
m1<-geem2(clearance~trt,id=id,data=keratosis,family=binomial,corstr="independence")
m2<-geem2(pain~trt,id=id,data=keratosis[keratosis$lesion==1,],family=gaussian,corstr="independence")
mmmgee(x=list(m1,m2),biascorr=TRUE)

# }