SLTCA: Scalable and Robust Latent Trajectory Class Analysis Using Artificial Likelihood

The package is on CRAN now. This repository will mainly serve as a platform for bug reporting (at Issues) while we will post the lastest version on CRAN. To access the latest version on GitHub, please go to branch CRAN. Thanks for considering using our software!

install.packages("SLTCA")
library(SLTCA)

Currently SLTCA supports R version >= 3.5.0.

By default, the function simulation(n) generates a dataset with n observations under the scenario 1 described by Hart, Fei and Hanfelt (2020).

# generate a dataset with 500 individuals
dat <- simulation(500)

Specifically, it returns a data frame of 2 latent classes with 6 longitudinal features y.1 to y.6, including count (y.1 and y.2), binary (y.3 and y.4) and continuous (y.5 and y.6) features. The data frame also consists of individual identifiers (id), corresponding time of longitudinal features (time) and the number of visit (num_obs). In addition, variable baselinecov is a binary baseline risk factor of latent classes. Variable latent is the true latent class labels.

The analysis for the dataset dat can be conducted by running SLTCA function:

res <- SLTCA(k=1,dat=dat,num_class=2,id="id",time="time",num_obs="num_obs",features=paste("y.",1:6,sep=''),
             Y_dist=c('poi','poi','bin','bin','normal','normal'),
             covx="baselinecov",ipw=1,stop="tau",tol=0.005,max=50,
             varest=T,balanced=T,MSC='EQIC',verbose=T)

Please refer to the function documentation for more details.

Hart, Fei and Hanfelt (2020), Scalable and Robust Latent Trajectory Class Analysis Using Artificial Likelihood. Biometrics, Accepted Author Manuscript. doi:10.1111/biom.13366