Skip to main content

Table 5 Model comparison with different distributional assumptions

From: Time to death and its determinants among under-five children in Sub-Saharan Africa using the recent (2010–2018) demographic and health survey data: country-based shared frailty analyses

Model Baseline hazard distribution Frailty distribution Frailty variance (θ, p-value) AIC BIC LLR
Cox-model NA Gamma     
Shard frailty Exponential Gamma 0.07, < 0.001 99,082.74 99,404.08 −49,510.37
Shard frailty Weibull Gamma 0.07, < 0.001 98,084.63 98,416.34 −49,010.32
Shard frailty Gompertz Gamma 0.07, < 0.001 96,693.04 97,024.74 −48,314.52
Shard frailty Log-logistic Gamma The model didn’t converge
Shard frailty Lognormal Gamma The model didn’t converge
  1. Model with inverse Gaussian frailty distribution did not converge
  2. AIC Akakian Information Criteria, BIC Bayesian Information Criteria, LLR Loglikelihood Ratio, NA Not Applicable