ANALISIS PERTAMBAHAN TINGGI BADAN BALITA STUNTING DI PROVINSI SUMATERA BARAT MENGGUNAKAN METODE REGRESI KUANTIL BINER BAYESIAN

Abstract

Stunting is one of the national health problems in Indonesia, where children experience growth failure. This study aims to construct a model for the classification of height gain of stunting toddlers in West Sumatra Province using the Bayesian binary quantile regression method. Quantile analysis is used in the Bayesian concept to produce more effective and natural estimates, especially for data that are not normally distributed. In the Bayesian approach, the Asymmetric Laplace Distribution (ALD) is used in determining the likelihood function as the basis for forming the posterior distribution used in the parameter estimation process. Bayesian binary quantile regression is an extension of quantile regression where the scale of the dependent variable used is binary. The research data used is data on 1000 stunting toddlers in West Sumatra Province in August 2021 and February 2022. In this study, it was found that the Bayesian binary quantile regression method at quantile 0.50 was the best conjecture model in classifying the level of height gain of stunting toddlers in West Sumatra Province. The criteria for model goodness are based on the greatest accuracy value. Factors that are significant in influencing the height gain of stunting toddlers in West Sumatra Province are birth weight, exclusive breastfeeding, and immunization

Keywords

• stunting, bayesian binary quantile regression method, asymmetric laplace distribution