College of Computer Science and Mathematics

Discussion of a master’s thesis in the College of Computer Science and Mathematics – Department of Statistics and Informatics entitled (Forecasting using symmetric patterns and equation error models)
A master’s thesis was discussed at the College of Computer Science and Mathematics at the University of Mosul on Tuesday, 4/26/2022, to predict using symmetric patterns and equation error models for the student Younis Muhammad Thanoun and under the supervision of Assistant Professor Dr. Osama Hayam Abdel Majid HayawiTime-series data are built the same as the various data in many fields of statistics such as regression analysis data and others. These data are subject to pollution through the presence of abnormal values ​​that affect the phenomenon under study and also affect the accuracy of the results that can be obtained even after processing the data in terms of stability In the mean and variance, it has been suggested to use a kind of filter that is used with time series data in order to filter it and prepare its readiness for analysis, so the researcher suggested using the wavelet filter to filter the data and then analyze it and find the predictive values ​​of the phenomenon under study using symmetrical patterns, which is a method of prediction suggested Since 2001By the scientist Singh and then using the wavelet filter data, specifically the wavelet Haar, to predict a set of stochastic linear kinetic models, which is a set of equation error models that includes two types of kinetic models, namely, the autoregressive model with exogenous inputs, symbolized by ARX, and the autoregressive model and averages Autoregressive and Moving Average with exogenous inputs, symbolized by ARMAX.The application in this thesis was done on data representing the turbidity percentage in the potable water after the filtration process, which was considered as an output variable, any variable resulting from the filtration process for potable water and denoted by Y_t, and the turbidity percentage in the potable water before the filtration process and denoted by U_t, as it was taken A sample of 135 observations was divided into two parts, the first part was for estimation and the second part was used as a dimensional sample in the comparison process with the predictive values ​​obtained from symmetric patterns and equation error models. The preference of kinetic models was reached using the wavelet in giving the best predictive values ​​for the phenomenon under study Relying on some statistical and engineering criteria.The discussion committee was chaired by Prof. Dr. Abdul Ghafour Jassim Salem and the membership of both Assistant Professor Dr. Dilshad Shaker Ismail from Salah El-Din University / College of Administration and Economics and Assistant Professor Dr. Osama Bashir Hanoun and under the supervision and membership of Assistant Professor Dr. Hiam Abdul Majeed Hayawi.