A Flexible Bayesian Parametric Proportional Hazard Model: Simulation and Applications to Right-Censored Healthcare Data

Table 8

Numerical summaries of posterior characteristics based on McMC sample for GLL PH model for the larynx cancer data.

Characteristics	Pars
	Alpha	(stage 2)	(stage 3)	(stage 4)	(age)	(diagyr)	Eta	Kappa
Mean	1.539	−0.182	0.376	1.222	0.187	−0.111	0.869	0.336
SD	0.215	0.454	0.337	0.411	0.144	0.149	0.247	0.077
Naïve SE	0.002	0.004	0.003	0.004	0.001	0.001	0.002	0.001
Time series SE	0.002	0.004	0.003	0.004	0.001	0.001	0.003	0.001
Minimum	0.847	−1.975	−0.902	−0.531	−0.373	−0.730	0.197	0.112
2.5^th percentile	1.157	−1.108	−0.289	0.396	−0.091	−0.403	0.457	0.207
Q1	1.389	−0.480	0.152	0.952	0.089	−0.212	0.691	0.282
Medium (Q2)	1.524	−0.170	0.377	1.230	0.187	−0.112	0.846	0.328
Q3	1.668	0.128	0.605	1.498	0.284	−0.012	1.020	0.382
97.5^th percentile	2.005	0.667	1.030	2.010	0.476	0.181	1.412	0.507
Maximum	2.701	1.648	1.770	2.848	0.817	0.509	2.131	0.763
Mode	1.550	−0.100	0.300	1.300	0.150	−0.150	0.850	0.325
Variance	0.046	0.207	0.113	0.169	0.021	0.022	0.061	0.006
Skewness	0.447	−0.173	−0.041	−0.086	0.081	0.023	0.595	0.604
Kurtosis	0.511	0.068	0.010	0.070	0.102	0.027	0.514	0.656
95% credible interval	(1.157, 2.005)	(−1.108, 0.667)	(−0.289, 1.030)	(0.396, 2.010)	(−0.091, 0.476)	(−0.730, 0.181)	(0.197, 1.412)	(0.112, 0.507)
P (>0|data)	1.000	0.352	0.870	0.998	0.906	0.227	1.000	1.000