A Proposed Approach for Joint Modeling of the Longitudinal and Time-To-Event Data in Heterogeneous Populations: An Application to HIV/AIDS’s Disease
Table 1
Comparison of the estimation of parameters in the three approaches in different sample sizes, .
Models
Class
Parameter
True value
Bias
SE
CP
Bias
SE
CP
Bias
SE
CP
JLCM
Class 1
Intercept
0.5
0.068
0.210
0.931
0.042
0.183
0.944
0.013
0.174
0.953
Class 2
Intercept
−0.5
0.053
0.199
0.929
0.031
0.187
0.929
0.011
0.176
0.958
PA
Class 1
Intercept
0.5
0.062
0.198
0.929
0.040
0.210
0.959
0.033
0.197
0.951
0
0.032
0.197
0.931
0.001
0.187
0.951
0.000
0.178
0.953
Class 2
Intercept
−0.5
0.037
0.212
0.927
0.029
0.211
0.951
0.027
0.196
0.952
0
0.021
0.194
0.927
0.000
0.191
0.949
0.000
0.167
0.955
SA
Class 1
Intercept
0.5
0.061
0.224
0.798
0.043
0.214
0.859
0.035
0.194
0.878
0
0.032
0.281
0.842
0.001
0.198
0.851
0.000
0.163
0.880
Class 2
Intercept
−0.5
0.037
0.228
0.782
0.031
0.213
0.810
0.029
0.187
0.894
0
−0.021
0.273
0.846
−0.000
0.179
0.829
−0.000
0.157
0.878
JLCM: joint latent class models; PA: proposed approach; SA: separate approach; SE: standard error; CP: coverage probability of the 95% confidence interval.