EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Modeling Stunting Cases in North Sumatra Using Generalized Poisson Regression with Maximum Likelihood Estimation

Elma Dwi Ariana Aprilia Zam, Fakhrur Rozi Nasution, Riri Syafitri Lubis

Desimal: Jurnal Matematika · 27.06.2026 · 9 (2) · 399-415

DOI: 10.24042/djm.v9i2.32190 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Desimal: Jurnal Matematika, online erschienen am 27.06.2026. Verfasst von Elma Dwi Ariana Aprilia Zam und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stunting remains a major public health challenge in Indonesia, particularly in regions where case distribution varies substantially across administrative areas. This study aimed to model the number of stunting cases in 33 regencies and municipalities of North Sumatra Province in 2023 using Generalized Poisson Regression estimated through Maximum Likelihood Estimation. A quantitative explanatory design was applied using secondary aggregate data obtained from the North Sumatra Provincial Health Office. The response variable was the number of stunting cases, while the explanatory variables included the number of low birth weight infants, pregnant women consuming iron supplementation tablets, pregnant women experiencing chronic energy deficiency, and infants receiving vitamin A supplementation. The analysis involved descriptive statistics, multicollinearity testing, Poisson regression, overdispersion assessment, Generalized Poisson Regression, parameter significance testing, incidence rate ratio interpretation, and model comparison using the Akaike Information Criterion. The Poisson model showed severe overdispersion, with a deviance-to-degree-of-freedom ratio of 525.357, indicating that the equidispersion assumption was violated. Generalized Poisson Regression provided a substantially better model fit, with an AIC of 505.8721 compared with 14985.5643 for Poisson regression. The final model identified chronic energy deficiency among pregnant women as the only statistically significant predictor of stunting cases (β = 0.42770, p = 0.0004; IRR = 1.5337; 95% CI: 1.2099–1.9442). These findings demonstrate that Generalized Poisson Regression is appropriate for overdispersed stunting count data and highlight maternal nutritional vulnerability as an important factor associated with regional variation in stunting cases in North Sumatra.

Zitieren

APA

Zam, E. D. A. A., Nasution, F. R., & Lubis, R. S. (2026). Modeling Stunting Cases in North Sumatra Using Generalized Poisson Regression with Maximum Likelihood Estimation. Desimal: Jurnal Matematika, 9(2), 399-415. https://doi.org/10.24042/djm.v9i2.32190

Vancouver

Zam EDAA, Nasution FR, Lubis RS. Modeling Stunting Cases in North Sumatra Using Generalized Poisson Regression with Maximum Likelihood Estimation. Desimal: Jurnal Matematika. 2026;9(2):399-415. doi:10.24042/djm.v9i2.32190

BibTeX
@article{Zam2026,
  author = {Zam, Elma Dwi Ariana Aprilia and Nasution, Fakhrur Rozi and Lubis, Riri Syafitri},
  title = {Modeling Stunting Cases in North Sumatra Using Generalized Poisson Regression with Maximum Likelihood Estimation},
  journal = {Desimal: Jurnal Matematika},
  year = {2026},
  volume = {9},
  number = {2},
  pages = {399--415},
  doi = {10.24042/djm.v9i2.32190}
}
Kostenlos starten →