residuals.glmgee: Residuals for Generalized Estimating Equations

In glmtoolbox: Set of Tools to Data Analysis using Generalized Linear Models

Residuals for Generalized Estimating Equations

Description

Calculates residuals for a fitted generalized estimating equation.

Usage

## S3 method for class 'glmgee'
residuals(
  object,
  ...,
  type = c("mahalanobis", "pearson", "deviance"),
  plot.it = FALSE,
  identify
)

Arguments

object	a object of the class glmgee.
...	further arguments passed to or from other methods
type	an (optional) character string giving the type of residuals which should be returned. The available options are: (1) "pearson"; (2) "deviance"; (3) the distance between the observed response vector and the fitted mean vector using a metric based on the product between the cluster size and fitted variance-covariance matrix ("mahalanobis"). By default, type is set to be "mahalanobis".
plot.it	an (optional) logical switch indicating if a plot of the residuals is required. By default, plot.it is set to be FALSE.
identify	an (optional) integer value indicating the number of individuals/clusters to identify on the plot of residuals. This is only appropriate when plot.it=TRUE.

Value

A vector with the observed residuals type type.

Examples

###### Example 1: Effect of ozone-enriched atmosphere on growth of sitka spruces
data(spruces)
mod1 <- size ~ poly(days,4) + treat
fit1 <- glmgee(mod1, id=tree, family=Gamma(log), data=spruces, corstr="AR-M-dependent")
### Plot to assess the adequacy of the chosen variance function
residuals(fit1, type="deviance", plot.it=TRUE, col="red", pch=20, col.lab="blue",
          col.axis="blue", col.main="black", family="mono", cex=0.8)
### Plot to identify trees suspicious to be outliers
residuals(fit1, type="mahalanobis", plot.it=TRUE, col="red", pch=20, col.lab="blue",
          col.axis="blue", col.main="black", family="mono", cex=0.8)

###### Example 2: Treatment for severe postnatal depression
data(depression)
mod2 <- depressd ~ visit + group
fit2 <- glmgee(mod2, id=subj, family=binomial(logit), corstr="AR-M-dependent", data=depression)
### Plot to identify women suspicious to be outliers
residuals(fit2, type="mahalanobis", plot.it=TRUE, col="red", pch=20, col.lab="blue",
          col.axis="blue", col.main="black", family="mono", cex=0.8)

###### Example 3: Treatment for severe postnatal depression (2)
mod3 <- dep ~ visit*group
fit3 <- glmgee(mod3, id=subj, family=gaussian(identity), corstr="AR-M-dependent", data=depression)
### Plot to assess the adequacy of the chosen variance function
residuals(fit3, type="pearson", plot.it=TRUE, col="red", pch=20, col.lab="blue",
          col.axis="blue", col.main="black", family="mono", cex=0.8)
### Plot to identify women suspicious to be outliers
residuals(fit3, type="mahalanobis", plot.it=TRUE, col="red", pch=20, col.lab="blue",
          col.axis="blue", col.main="black", family="mono", cex=0.8)

glmtoolbox documentation built on Oct. 10, 2023, 9:06 a.m.