The goal of gscaLCA is to execute Latent Class Analysis by using Generalized Structure Component Analaysis(GSCA).
You can install the released version of gscaLCA from CRAN with:
And the development version from GitHub with:hee6904/gscaLCA
This is a basic example.
library(gscaLCA)
#> Loading required package: gridExtra
#> Loading required package: ggplot2
#> Loading required package: stringr
#> Loading required package: progress
#> Loading required package: psych
#>
#> Attaching package: 'psych'
#> The following objects are masked from 'package:ggplot2':
#>
#> %+%, alpha
#> Loading required package: fastDummies
#> Loading required package: fclust
#> Loading required package: MASS
#> Loading required package: devtools
#> Loading required package: usethis
#> Loading required package: doSNOW
#> Loading required package: foreach
#> Loading required package: iterators
#> Loading required package: snow
#> Loading required package: doParallel
#> Loading required package: parallel
#>
#> Attaching package: 'parallel'
#> The following objects are masked from 'package:snow':
#>
#> clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
#> clusterExport, clusterMap, clusterSplit, makeCluster,
#> parApply, parCapply, parLapply, parRapply, parSapply,
#> splitIndices, stopCluster
## basic example codeRunning gscaLCA AddHealth data. The number of cluster is 2 and the number of bootstrap is 20.
R2 = gscaLCA(AddHealth, varnames = names(AddHealth)[2:6], num.cluster = 2)
#> Warning in gscaLCA(AddHealth, varnames = names(AddHealth)[2:6], num.cluster
#> = 2): Listwise delection was used. Uncompleted data is not available in the
#> current version
#> =========================================================
#> LCA by using Fuzzing Clusterwise GSCA
#> =========================================================
#> Fit for 2 latent classes:
#> number of used observations: 5066
#> number of deleted observation: 48
#> number of bootstrap for SE: 20 / 20
#>
#> MODEL FIT -----------------------------------------------
#> FIT : 0.9997
#> AFIT : 0.9997
#> FPI : 0.6818
#> NCE : 0.7504
#>
#> Estimated Latent Class Prevalnces (%) -------------------
#> 44.91% 55.09%
#>
#> Conditional item response probability -------------------
#> $Smoking
#> Class Category Estimate
#> 1 Latent Class 1 Yes 0.9653
#> 2 Latent Class 1 No 0.0347
#> 3 Latent Class 2 Yes 0.3956
#> 4 Latent Class 2 No 0.6044
#>
#> $Alcohol
#> Class Category Estimate
#> 1 Latent Class 1 Yes 0.9684
#> 2 Latent Class 1 No 0.0316
#> 3 Latent Class 2 Yes 0.6686
#> 4 Latent Class 2 No 0.3314
#>
#> $Drug
#> Class Category Estimate
#> 1 Latent Class 1 Yes 0.4497
#> 2 Latent Class 1 No 0.5503
#> 3 Latent Class 2 Yes 0.0262
#> 4 Latent Class 2 No 0.9738
#>
#> $Marijuana
#> Class Category Estimate
#> 1 Latent Class 1 Yes 0.9754
#> 2 Latent Class 1 No 0.0246
#> 3 Latent Class 2 Yes 0.1985
#> 4 Latent Class 2 No 0.8015
#>
#> $Cocaine
#> Class Category Estimate
#> 1 Latent Class 1 Yes 0.3956
#> 2 Latent Class 1 No 0.6044
#> 3 Latent Class 2 Yes 0.0254
#> 4 Latent Class 2 No 0.9746