Table 4.

Results from simulation study under setting 2

Type of parameterb		True value	Mean estimate	Sample SEa	Mean Asy. SEa	Mean bootstrap SE
β12	Age	0.036	0.037	0.0060	0.0016	0.0062
	Tran	0.16	0.15	0.042	0.035	0.039
	Frail	0.90	0.87	0.190	0.120	0.190
	Frail:tran	−0.084	−0.081	0.039	0.020	0.040
	Alone	−0.26	−0.28	0.096	0.074	0.099
	Alone:tran	0.037	0.038	0.031	0.024	0.030
	Depress	0.28	0.25	0.094	0.053	0.095
β13	Age	0.075	0.075	0.020	0.003	0.019
β21	Age	−0.015	−0.015	0.0054	0.003	0.0056
	Tran	−0.012	−0.015	0.012	0.019	0.017
	Frail	−0.26	−0.27	0.17	0.14	0.17
β23	Frail	−0.81	−0.82	0.27	0.26	0.27
	Tran	−0.13	−0.12	0.066	0.066	0.069
	Frail:tran	0.12	0.13	0.057	0.057	0.062
	Sex	−0.81	−0.81	0.18	0.19	0.19
γ12		1	1	0	0	0
γ13		−0.5	−0.66	1.30	1.07	1.32
γ21		1	1	0	0	0
γ23		5.0	5.36	1.79	1.44	1.82
Variance	var(ηi1)	0.16	0.13	0.050	0.052	0.049
Components	cov(ηi1, ηi2)	−0.05	0.036	0.013	0.016	0.012
	var(ηi2)	0.02	0.015	0.006	0.008	0.006

^aThe “Sample SE” denotes the standard deviation of the parameter estimates from 100 simulated data sets and the “Mean Asy. SE” denotes the average of the asymptotic standard errors obtained from the 100 simulated data sets.

^bWe use k to index the state from which transition is made. We use l = 2 and 3 to denote the states of disability and death that can be transited to from state 1. We use l = 1 and 3 to denote the states of independence and death that can be transited to from state 2.