predict.SMNCens: Prediction of future observations from an smn.clmm object

In skewlmm: Scale Mixture of Skew-Normal Linear Mixed Models

Prediction of future observations from an smn.clmm object

Description

Predicted values are obtained through conditional expectation. For details, see Schumacher, Lachos, and Matos (2021).

Usage

  ## S3 method for class 'SMNclmm'
predict(object, newData, ...)

Arguments

object	An object inheriting from class SMNclmm, representing a fitted scale mixture normal linear mixed model.
newData	A data frame for which response variable should be predicted, including covariates, groupVar and possibly timeVar. If missing or NULL, fitted values are returned.
...	Additional arguments.

Value

A data frame with covariates, groupVar and ypred, where ypred contains the predicted values from the response variable.

Author(s)

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

References

Schumacher, F. L., Lachos, V. H., and Matos, L. A. (2021). Scale mixture of skew-normal linear mixed models with within-subject serial dependence. Statistics in Medicine 40(7), 1790-1810.

See Also

smn.clmm, fitted.SMNclmm

Examples

# Simulated example: 20 individuals with 5 times for estimation and
# 1 time for prediction
set.seed(963)
nj1 = 6; m = 20
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=0.4,
                  D=0.6*diag(1), beta=c(1,2), depStruct="CS", phi=0.25)
# Estimation
fm1 = smn.clmm(subset(dat1,time<6), formFixed=y~x, groupVar="ind",
               depStruct="CS", ci="ci", lcl="lcl", ucl="ucl",
               control=lmmControl(max.iter=30, tol=1e-4))
# Prediction
pred = predict(fm1, subset(dat1,time==6))
# MSPE
mean((subset(dat1,time==6)$y - pred$ypred)^2)

skewlmm documentation built on July 9, 2023, 7:29 p.m.