A Comparative Study between Logistic Regression and Neural Networks for Examining Factors Influencing Child Mortality in Libya. | ||||
The Egyptian Statistical Journal | ||||
Article 1, Volume 64, Issue 2, December 2020, Page 1-26 PDF (713.1 K) | ||||
Document Type: Original Article | ||||
DOI: 10.21608/esju.2020.189427 | ||||
![]() | ||||
Authors | ||||
El-Sayed Khater1; Sayada Abdel Nabi2; Mohamed Abdel Kader3; Samira Omer Eldofani4 | ||||
1Department of Biostatistics and Demography, Faculty of Graduate studies for Statistical Research, Cairo University, Egypt | ||||
2Faculty of Commerce, Al-Azhar University – (Girls' Branch), Egypt | ||||
3Faculty of Commerce, Al-Azhar University – (Boys' Branch), Egypt | ||||
4PhD student, Libya | ||||
Abstract | ||||
Child mortality remains one of the most critical issues under investigation. This study focuses on describing the phenomenon of child mortality in Libya and examining the factors influencing it. Since child mortality is a variable that follows a Bernoulli process, logistic regression is utilized to estimate the model that represents the relationship between the variables and employs it for statistical prediction. Logistic models also enable prediction of the occurrence or non-occurrence of specific events. However, certain assumptions may not hold true, prompting the use of neural networks. Neural networks have the capability to model data and make predictions without relying on specific assumptions about the nature of the variables or their relationships. | ||||
Keywords | ||||
Neural Networks; Logistic regression; Infant mortality; Libya | ||||
Statistics Article View: 430 PDF Download: 557 |
||||