plot.mahalDist.SMNCens function

Plot Mahalanobis distance for a fitted smn.clmm

Plot method for objects of class "mahalDistCens". It also gives a quantile for outlier detection, based on the Mahalanobis distance theoretical distribution.

## S3 method for class 'mahalDistCens'
plot(x, fitobject, level = 0.99, nlabels = 3, ...)

Arguments

• x: An object inheriting from class mahalDistCens, representing the Mahalanobis distance from a fitted scale mixture of normal censored linear mixed model.
• fitobject: Optional. An object inheriting from class SMNclmm, representing the fitted scale mixture of normal linear mixed model that was used for calculating the Mahalanobis distance.
• level: An optional numeric value in $(0,1)$ indicating the level of the quantile. Default is 0.99.
• nlabels: Number of observations that should be labeled. Default is 3.
• ...: Additional arguments.

Returns

A ggplot object, plotting the index versus the Mahalanobis distance, if all subject have the same number of observations; or plotting the number of observations per subject versus the Mahalanobis, otherwise.

Author(s)

Fernanda L. Schumacher, Larissa A. Matos, Victor H. Lachos and Katherine L. Valeriano

See Also

ggplot , mahalDistCens , smn.clmm

Examples

nj1 = 5; m = 30
time = rep(1:nj1, times=m)
groups = as.factor(rep(1:m, each=nj1))
dat1 = rsmsn.clmm(time, groups, cbind(1,time), rep(1,m*nj1), sigma2=0.7,
                  D=0.5*diag(1), beta=c(1,2), depStruct="CS", phi=0.4)
# Estimation
fm1 = smn.clmm(dat1, formFixed=y~x, groupVar="ind", depStruct="CS", ci="ci",
               lcl="lcl", ucl="ucl", control=lmmControl(max.iter=30))
distance = mahalDistCens(fm1)
plot(distance, level=0.95, nlabels=2)