A Study On The Application Of International Labour Office's (ILO) Financing Models And Distribution Fitting Methods In The Actuary Of Social Health Insurance

Objectives To explore the feasibility and value of application of ILO financing models in actuary of social health insurance. To compare the performance of distribution fitting methods, and apply the best method to the expenditure estimation model of ILO financing models to calculate the contribution rate of some social health insurance scheme with true data.Methods One-year actuary models which were developed from the ILO financing models were used to calculate the contribution rate. Gamma distribution, Weibull distribution, Pareto distribution, log-normal distribution, log-logisitc distribution, Pearson-V distribution, Inv-Gaussian distribution and Kernel Density Estimation (KDE) were used to fit the distribution of claim data and chi-square test was used to test the goodness-of-fit. Sheather and Jones Plug-In approach (SJPI) was used to select the optimal bandwidth of KDE and standard normal function was used as a kernel function fo KDE.Results Results of comparing the goodness-of-fit tests of different distribution fitting methods showed that KDE was thet best distribution fitting method. The mean of every claim estimated by KDE was applied to one-year actuary models in order to compute the contribution rate of Medical insurance scheme of hospitalization in Chengdu's urban areas. The result of calculation was 4.7%.Conclusion Research of actual data showed that the application of ILO financing models in actuary of social health insurance was feasible and ILO financing models was comprehensive and adjustable. It was proved that KED was suitable for fitting distribution of claim data and the combination of ILO financing models and KED provide a new way in actuary of social health insurance.