Simple Regression Models: A Comparison using Criteria Measures

Publication Date: 09/05/2024

DOI: 10.52589/AJMSS-VKLVNUU5


Author(s): Osuagwu Chidimma Udo, Okenwe Idochi.

Volume/Issue: Volume 7 , Issue 2 (2024)



Abstract:

The study is on simple regression models: a comparison using criteria measures. The source of the dataset used for this study was extracted from records of the Federal Medical Centre, Owerri, Imo State, on weight of babies and hemoglobin level of mothers. The response variable is weight of babies while the explanatory variable is hemoglobin level of mothers. Eleven simple regression models—Linear, Growth, Quadratic, Polynomial, Logarithmic, Hyperbolic, Power, Exponential Growth, Square Root, Sinusoidal and Arctangent—were stated and employed for the study. For ease of data analysis, E-views package was implemented. Three model selection criteria measures for comparison, known as Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC) and Hannan-Quinn Information Criterion (HQIC), were employed. The result of the study showed that, when it comes to analyzing the association between baby weight and mothers' hemoglobin levels, the exponential growth regression model performs better than the other ten models that were examined. Therefore, researchers should investigate other models that were not included in this analysis and compare the findings using goodness of fit metrics other than the criteria measures used in this work.


Keywords:

Simple Nonlinear Regression, Simple Linear Regression, AIC, SIC, HQIC, Model Comparison.


No. of Downloads: 0

View: 100




This article is published under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
CC BY-NC-ND 4.0