lcmm (version 1.6.6)

link.confint: Confidence intervals for the estimated link functions from lcmm and multlcmm

Description

This function provides 95% confidence intervals around the estimated transformation given in estimlink attribute of lcmm and multlcmm objects. It can also be used to evaluate the link functions at other values than those given in attribute estimlink of lcmm or multlcmm object. Furthermore, a graph of the link functions, representing the relation between the underlying latent process and the observed outcomes, and the confidence intervals can be plotted.

Usage

## S3 method for class 'lcmm':
link.confint(x,ndraws=2000,Yvalues,plot=TRUE,...)
## S3 method for class 'multlcmm':
link.confint(x,ndraws=2000,Yvalues,plot=TRUE,...)

Arguments

x

an object inheriting from classes lcmm or multlcmm.

ndraws

the number of draws that should be generated to approximate the posterior distribution of the transformed values. By default, ndraws=2000.

Yvalues

a vector (for a lcmm object) or a matrix (for a multlcmm object) containing the values at which to compute the transformation(s). Default to the values in x$estimlink.

plot

logical indicating if the link functions and confidence intervals should be plotted. Default to TRUE.

...

other parameters to be passed through to plotting functions

Value

• For a lcmm object, returns a data frame containing the values at which the transformation is evaluated, the transformed values and the lower and the upper limits of the confidence intervals.

  For a multlcmm object, returns a data frame containing the indicator of the outcome, the values at which the transformations are evaluated, the transformed values and the lower and the upper limits of the confidence intervals.

See Also

lcmm, multlcmm, plot.linkfunction

Examples

data(data_Jointlcmm)

## Univariate mixed model with splines link funciton
m14<-lcmm(Ydep2~Time+I(Time^2),random=~Time,subject='ID',ng=1,
data=data_Jointlcmm,link="5-manual-splines",intnodes=c(10,20,25))
##Transformed values of several scores and their confidence intervals
transf.m14 <- link.confint(m14,ndraws=2000,Yvalues=c(0,1,7:30),plot=TRUE)

## Multivariate mixed model with splines link functions
m1 <- multlcmm(Ydep1+Ydep2~1+Time*X2+contrast(X2),random=~1+Time,
subject="ID",randomY=TRUE,link=c("4-manual-splines","3-manual-splines"),
intnodes=c(8,12,25),data=data_Jointlcmm,
B=c(-1.071, -0.192,  0.106, -0.005, -0.193,  1.012,  0.870,  0.881,
  0.000,  0.000, -7.520,  1.401,  1.607 , 1.908,  1.431,  1.082,
 -7.528,  1.135 , 1.454 , 2.328, 1.052))
##Confidence intervals for the transformed values (given in m1$estimlink)
transf.m1 <- link.confint(m1,ndraws=2000,plot=TRUE)