Using the Box-Jenkins methodology to predict the number of participants for the Continuing Education Center at the Middle Technical University

Abstract

            The time series analysis method is one of the important statistical methods in studying a phenomenon over a certain period. Through this analysis, the pattern of the series is identified and used to predict the future values ​​of the phenomenon, which helps in making the appropriate decision and developing sound plans for the future. The research aims to choose the best model to predict the number of participants for the Continuing Education Center at the Central Technical University, using Jenkins Box models, and to choose the best model by comparing models by using differentiation criteria, as the models developed by the two scientists were relied upon, (Jenkins & Box Jenkins) and it is during the development of a methodology in the study and analysis of autoregressive models - moving media for time series starting from the process of checking the stability through the autocorrelation and partial autocorrelation integrated ARIMA, the instability of the series appeared and stabilized after taking the first difference, and to ensure the significance of the stability, the ADF test was relied on. Determining the best model based on the used accuracy criterion (AIC). The greatest possibility function was used in estimating the model parameters, where the ARIMA model (1,1, 0) was chosen as the best model from the Box Jenkins models, and the significance of the chosen model was tested for the studied phenomenon to ensure the suitability of the model from During the confidence interval test for the residuals, which indicates that the error series of the model is random, as well as the plot of the original series And the estimated series of the estimated model ARIMA(1,1,0), a semi-match was observed between the curves of the original series and the estimated, and this gives us an idea of ​​the importance and accuracy of the estimated model (ARIMA(1,1,0). One of the most important conclusions is that the original series was unstable When the Dickey-Fuller test was conducted, it became stable when the first difference was taken, and one of the most important recommendations is to take advantage of the results of the research to develop appropriate plans and policies in developing the work of the Continuing Education Center, preparing the training rooms and equipping them with modern technologies, developing and diversifying training curricula and improving other services in the center. In the center in terms of training and engaging them in training and development courses in the field of (computer and administrative development) to improve their performance to meet the changes in the number of participants. The statistical program Eviews was used.