Abstract:
: This study aimed to compare the parametric survival models such as exponential model, Weibull model, Log-logistic model, lognormal model and Gompertz model to determine the best fits model to estimate and evalute the service provided from health insurance to the citizen in khartoum region subscriptions that benefits from service provided from hospital or pharmacies. The Models were estimated by using maximum likelihood method, and Log-likelihood and Akaike Information Criterion (AIC), paysian criterion (BIC) were used as selection criterion methods. The results of AIC and Log-likelihood show that Log-logistics regression model best fit the the servise provided from health insurance to the citizen. The results of exponential regression model revealed that revenues and benefits are prognostic factors affecting the the service that provided from health insurance to the citizen. Therefore, it is recommended that researchers of the health care unit consider this model (exponential regression model) in their researches concerning other field.
|