7 December، 2022
master’s thesis entitled: Estimation and Variable selection in Balanced and unbalanced Panel data model
Discussion of a master’s thesis in the College of Computer Science and Mathematics – Department of Statistics and Informatics entitled:Estimation and Variable selection in Balanced and unbalanced Panel data modelIn continuation of the scientific research movement and with the follow-up and presence of the Dean of the College of Computer Science and MathematicsProf. Dr. Duha Bashir Abdullah The esteemedIt was discussed in the discussion room at the Faculty of Computer Science and Mathematics at the University of Mosul on Tuesday 7/12/2022Master’s thesis by studentAbeer Adel Hameed AL-Jber, under the supervisionof ZakariyaYahyaALGamalProf.D.This study tookPropose nature-inspired algorithms represented by the black hole algorithm to estimate parameters and select explanatory variables in three regression models: the aggregate regression model, the fixed effects regression model, and the random effects regression model. These methods were applied to real data covering the fourteen Arab countries as longitudinal data for the time period (2000-2020) with a sample size of 294 observation to study the factors affecting the general index of stock prices. The results showed that the proposed method is superior to other traditional methods in terms of giving good estimates of the parameters of longitudinal data regression models based on the mean squares of errorThe study touchedIn many longitudinal data analysis studies, a large number of explanatory variables are usually studied, but sometimes the presence of such a large number of explanatory variables reduces the efficiency of longitudinal data models, and this is because a number of these explanatory variables may not be significant compared to other explanatory variables, which in turn leads to a decrease in the efficiency of forecasting. The problem of selecting explanatory variables is one of the important problems experienced by regression models in general and longitudinal data models in particular.This study aims To employ an intelligence technique, the Black Hole Algorithm (BHA), in selecting important independent variables in longitudinal data models in order to be more efficient compared to other variables selection methods by applying them to economic data.The discussion committee consists of:
| Prof. Dr.safayounistaleaa- Chairman |
| Assis. Prof. Dr. muzahimmohameedyahya- member |
| Lecturer. Dr. sadiqawadkazim ministry of higher education and scientific research- member |
| Prof. Dr. ZakariyaYahyaALGamal.– member and supervisor
Contact the university 
|
