Discussion of a higher diploma thesis in the College of Computer Science and Mathematics - Department of Statistics and Informatics entitled (Comparison of prediction between ARIMA and the random forest model)
On Sunday, 26-6-2022, the College of Computer Science and Mathematics at the University of Mosul discussed the Higher Diploma Thesis (Comparison of Prediction between ARIMA and the Random Forest Model) for the student (Uday Zaki Gerges) and under the supervision of Assistant Professor Dr. Osama Bashir Shukr Al-Hanun

The study dealt with the prediction of the quantities of agricultural evaporation and minimum temperatures as single time series data to diagnose the mathematical relationship of the autoregression of those variables. Iraqi data from one of the agricultural meteorological stations in the city of Mosul - Iraq were used as real study data. The study data faced many obstacles such as non-linearity, which will represent the most important reasons for the inaccuracy of predictions.
The study dealt with the use of the ARIMA season-hit model as a statistical method commonly used to model single time series and to obtain predictions of agricultural evaporation and minimum temperature variables. The random forest model was also used as one of the alternative methods for predicting time series based on the autoregressive principle.

The aim of the thesis is to use a random forest model to improve the prediction of agricultural evaporation and minimum temperature variables. Comparisons of the predictive results of agricultural evaporation and minimum temperatures showed the superiority of the random forest method over the traditional ARIMA method. Therefore, it can be concluded that the random forest method can be used as the best predictive method as an alternative method for the ARIMA model in classifying the evaporation and temperature data, which will lead to more accurate predictive results.
The discussion committee was chaired by the assistant professor, Dr. Muzahim Muhammad Yahya, and the membership of the teacher, Dr. Omar Salem Ibrahim, and the supervision and membership of the assistant professor, Dr. Osama Bashir Shukr Al-Hanun.

Also Read