Efficiency of Multiple Linear Regression in Identifying Factors Affecting Glycated Hemoglobin (HbA1c) Using R Programming. Enttsar Ali Arhema

كفاءة نموذج الانحدار الخطي المتعدد في تحديد العوامل المؤثرة في معدل السكر التراكمي

Abstract

This study aimed to analyze the relationship between glycated hemoglobin (HbA1c) and fasting blood sugar (FBS), age, and weight using a multiple linear regression model implemented in R programming language. The study adopted a descriptive-analytical approach and verified the key assumptions of the regression model, including normality, independence, multicollinearity, and absence of outliers.

           Statistical tests, including the Shapiro–Wilk test, Variance Inflation Factor (VIF), and Durbin–Watson test, confirmed that the model assumptions were satisfied. The results indicated that the regression model was statistically significant (F = 12.55, p < 0.05) and explained 45% of the variance in HbA1c levels.

           Furthermore, the findings showed that fasting blood sugar and age had a statistically significant effect on HbA1c levels, whereas weight did not show a significant effect. The study recommends expanding future analyses to include additional explanatory variables and highlights the importance of multiple linear regression and R programming in medical research.