International Journal of Statistics and Applied Mathematics
2021, Vol. 6, Issue 6, Part A
Multiple regression model selection via birth weight, mother age and gestation variablesAuthor(s):
Obaji Ifeoma and Nwagor PetersAbstract:
This study is on multiple regression model selection. The source of data is from records of 2019 to 2020 live deliveries in Federal Medical Center (FMC) Umuahia Abia State, Nigeria. The hospital keeps records of deliveries, ages of mothers and their birth weights. The mothers’ records that do not have the exact record of last menstrual period were not used for this study. Miscarriages i.e., pregnancies that did not go beyond 28 weeks gestation were not included in the study. It is a retrospective study of 100 pregnancies with outcomes of live births that exceeded 28 weeks gestation. The dependent variable is birth weight, while the independent variables are mother’s age and gestation. Four regression models; Lin-Log, Linear, Inverse, and Polynomial were examined in this study. The E-views software was used in this study. Four model selection techniques known as; coefficient of determination, Akaike Information Criterion, Schwarz Information Criterion, and Hannan-Quinn Information Criterion were used to select the best model. From the analysis in the overall goodness of fit assessment, the study concluded that the Lin-Log regression model performs slightly better than the other three regression models used in this study. Therefore, future researchers should look at a similar work by incorporating other nonlinear regression models like Double-Log and Log-Lin Regression models to compare results.Pages: 83-90 | Views: 181 | Downloads: 10Download Full Article: Click Here
How to cite this article:
Obaji Ifeoma, Nwagor Peters. Multiple regression model selection via birth weight, mother age and gestation variables. Int J Stat Appl Math 2021;6(6):83-90.