Determine the factors affecting the time to recovery of children with bacterial meningitis at Jigjiga university referral hospital in the Somali Regional State of Ethiopia: using the parametric shared frailty and AFT models

Table 3 Compares AFT and Frailty models using data on time to recovery status

Distribution	AIC	BIC	log-likelihood
Exponential	29672.82	29779.60	− 11148
Weibull	20063.76	20178.13	− 11256
Log logistic	20063.76	20178.13	− 11256
Generalized gamma	19883.78	20005.78	− 11126
Log-normal	*19865.06*	19997.29	− 11025

Baseline hazard function	Frailty distribution	AIC	BIC	DIC
Exponential	Gamma	3324.760	418.870	371.710
Exponential	Inverse Gaussian	3325.506	389.378	350.024
Weibull	Gamma	2824.988	392.919	353.197
Weibull	Inverse Gaussian	2813.650	394.808	354,871
Lognormal	Gamma	2691.597	410.978	379.897
Lognormal	Inverse Gaussian	2674.080	314.987	332.510
Log logistic	Gamma	2692.105	399.389	379.024
Log logistic	Inverse Gaussian	2675.287	389.991	396.179

Bold values indicate selected model that means Inverse Gaussian models was the appropriated for our study
AIC Akaike’s information criteria, BIC Bayesian information criterion, DIC deviance information criterion

ISSN: 1756-0500