Striking A Balance Between Contributions and Compensation for the Social Security Fund in Iraq

Abstract

Time series have gained great importance and have been widely applied in various fields such as economic, financial, health and social fields, which is a set of observations for successive time periods used in the analysis by studying the changes in the phenomenon and predicting the future of the phenomenon. The most important time series models are the black box models “ARIMAX” which is a mixed model consisting of self-regression with moving averages with external inputs and this gives the model a high accuracy. The process of building the model consists of several stages which are determining the model’s rank and the process of estimating the model’s parameters and then the process of predicting the condition of the fund to know the size of compensation granted to workers in the future and the possibility of balancing contributions and compensation in order to fulfill the fund's future obligations. The rank of the model was determined according to the criteria, AIC, MDL, FPE and estimating the model using ordinary least square method (OLS) and recursive least square (RLS) and also proposing a way to improve the prediction results by estimating the error and inserting it in a regression model as a factor influencing the prediction of the outputs to obtain the best predictable way for the future is by using data taken from the Social Security System for workers and from the period 1/2013 to 6/2019, where the proposed model outperformed the regular least squares method because it gave accurate results close to the real values.