Skip to main content

Table 5 Simulations results: class identification accuracy, calculated as the rate of correctly predicted class memberships, according to the number of individuals, n, and to the censoring rate, τ. The difference between the rates of the high and the low separation settings is provided

From: Joint latent class model: Simulation study of model properties and application to amyotrophic lateral sclerosis disease

n τ High separation Low separation Difference
100 5 0.9760 0.9418 -0.0342
  10 0.9767 0.9347 -0.0420
  15 0.9748 0.9248 -0.0500
  25 0.9689 0.9039 -0.0650
  50 0.9556 0.8335 -0.1221
500 5 0.9790 0.9440 -0.0350
  10 0.9778 0.9376 -0.0402
  15 0.9764 0.9321 -0.0443
  25 0.9720 0.9148 -0.0572
  50 0.9586 0.8458 -0.1128
1000 5 0.9814 0.9477 -0.0337
  10 0.9798 0.9419 -0.0379
  15 0.9782 0.9354 -0.0428
  25 0.9745 0.9186 -0.0559
  50 0.9605 0.8488 -0.1017
5000 5 0.9817 0.9480 -0.0337
  10 0.9801 0.9417 -0.0384
  15 0.9784 0.9348 -0.0436
  25 0.9748 0.9189 -0.0559
  50 0.9618 0.8504 -0.1114