| dc.description |
The ability to model and perform decision modeling and analysis is an essential feature of many real-world applications. Almost all the decisions in any organization are based on forecasts. Every decision becomes operational at some point in the future, so it should be based on forecasts of future condition. Keeping this in view, an attempt has been made in this dissertation work to study the pattern of NHIF memberships, to select the best model and to forecast the NHIF membership on monthly basis using statistical time series modeling.
National Health Insurance Fund membership enrolment forecasting is very essential in the management and providing good and quality health services to the citizen. Majority of decision makings like introduction of programs and infrastructure improvements are easily made with the aid of NHIF membership forecasts. In this dissertation, the NHIF membership enrolment of Dodoma region from 2002 to 2016 has been modeled using Box-Jenkins Autoregressive Integrated Moving Average model technique. The NHIF membership has been further forecasted to the financial year 2016/2017 to 2017/2018 (twenty four months) using seasonal ARIMA(1,2,0)(0,0,1)12 for male, seasonal ARIMA(0,2,0)(1,0,0)12 for female and seasonal ARIMA(0,2,1)(0,0,1)12 for total combined population. The model has been validated using Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Ljung-Box statistic values, graphical techniques like time series plots, Q-Q plots and histograms and p-values. The results show that the NHIF membership registration will grow up by the financial year 2016/2017 to 2017/2018 when only time is considered as a factor. The study also explores briefly the recourses required to support current and forecasted NHIF members. |
|