Skip to main content
. 2021 Oct 4;21:433. doi: 10.1186/s12887-021-02895-7

Table 4.

Model comparison with different distributional assumptions

Model Baseline hazard distribution Frailty distribution Frailty variance
(θ, p-value)
AIC BIC LRR
Cox-model NA Gamma 0.05, < 0.001 239,411 239,609 − 119,683
Shard frailty Exponential Gamma 0.06, < 0.001 76,330.47 76,247.13 −38,141.24
Shard frailty Weibull Gamma 0.06, < 0.001 76,168.05 76,393.73 −38,059.24
Shard frailty Gompertz Gamma 0.06, < 0.001 76,017.40 76,243.08 −37,983.70
Shard frailty Log-logistic Gamma The model did not converge
Shard frailty Lognormal Gamma

NB: Model with inverse Gaussian frailty distribution did not converge

NB: AIC = Akakian Information Criteria, BIC = Bayesian Information Criteria, LLR = Loglikelihood Ratio, NA = Not Applicable