evaluateGEEModel is used to get a generalized estimating equation of the
data by the adaptive shrinkage estimate method.
evaluateGEEModel(family, corstr, y, x, clusterID, criterion = "QIC",
theta = 0.75, gamma = 1, leastVar = 3, mostVar = ncol(x))A description of the error distribution and link function to be
used in the model. See family for details of family
functions.
A character string specifying the correlation structure. The following are permitted: "independence", "exchangeable" and "ar1".
The response data.
A data frame contains the covariate vectors.
The id for each subject in the initial samples. Note that the subjects in the same cluster will have identical id.
The model selection criteria, one of the 'PWD' or 'QIC'.
The parameters of the adaptive shrinkage estimate.
The parameters of the adaptive shrinkage estimate.
The minimum number of variables.
The maximum number of variables.
a list containing the following components
the correlation coefficient of the clusters
parameters that we estimate under the current samples
the sandwich information matrix for covariance
the index of the non zero coefficients
a list containing several matrices including the sandwich matrix
evaluateGEEModel fits the current data by generalized estimating equations(GEE) according to the value of the family argument and the corstr argument. We should notice that this is not the ordinary generalized estimating equations. It can determine the variables that have an impact on the response which called effective variables. We use model selection criteria like the QIC criterion to choose the optimal value.