Application of Poisson's Hierarchical Regression Model to the Deaths of Covid-19 in Mosul City Hospitals

• Published in: المجلة العراقية للعلوم الاحصائية Vol. 19; no. 2; pp. 1 - 21
• Main Authors: Ban Al ani, Mahmmood Altai
• Format: Journal Article
• Language: Arabic; English
• Published: College of Computer Science and Mathematics, University of Mosul 01-12-2022
• Subjects: fixed and random effects; full maximum likelihood method; hierarchical poisson regression model with random intercept; intraclass correlation coefficient
• ISSN: 1680-855X; 2664-2956
• DOI: 10.33899/iqjoss.2022.176199

This research deals with the study of the partial hierarchical Poisson regression model (with a random intercept), where this model is one of the most important models widely applied in analyzing data that is characterized by the fact that the observations take a hierarchical form. Where it the full maximum likelihood (FML) method is used to estimate the model parameters. The model was applied to the covid-9 deaths in Mosul city, were recorded during the period (1/1/202 - 1/9/2021), where four major hospitals in the city were selected to represent the group of second level of data (Ibn Sina Hospital, Al Salam Hospital, Shifa Hospital, General Mosul Hospital).
*طالب دبلوم عالی/ قسم الاحصاء والمعلوماتیة/ کلیة علوم الحاسوب والریاضیات/ جامعة الموصل.
**استاذ مساعد/ قسم الاحصاء والمعلوماتیة/ کلیة علوم الحاسوب والریاضیات/ جامعة الموصل.
تاریخ استلام البحث: 4/3/2022          تاریخ القبول: 9/4/2022            تاریخ النشر: 1/12/2022
The research found the adequate of the model for this type of data, as it was found that there are some factors that contribute to the increase in the number of deaths in the epidemic, such as the advanced age of the patient, the length of stay in the hospital, the percentage of oxygen in the patient's blood, in addition to the incidence of some chronic diseases such as asthma. The study recommended a more in-depth study of other types of these models, and the use of other estimation methods, in addition to paying attention to the methods of data recording by the city health department.