Pemodelan Prevalensi Balita Wasted di Pulau Jawa Menggunakan Multivariate Adaptive Regression Spline (MARS)

Hugo, Ben (2024) Pemodelan Prevalensi Balita Wasted di Pulau Jawa Menggunakan Multivariate Adaptive Regression Spline (MARS). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201070-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (3MB) | Request a copy

Abstract

Kekurangan gizi pada balita dapat menyebabkan kerusakan yang bersifat irreversible. Adapun wasting, yang juga dikenal sebagai malnutrisi akut, muncul akibat kekurangan asupan makanan berkualitas dan sering kali disertai oleh serangan berulang penyakit seperti diare, campak, dan malaria. Wasting dini juga meningkatkan risiko terhambatnya pertumbuhan di kemudian hari, termasuk risiko mengalami wasting dan stunting secara bersamaan, sehingga meningkatkan risiko kematian. Pada tahun 2022, prevalensi wasting pada balita di Indonesia sebesar 7,7%, yang mana belum mencapai target Sustainable Development Goals (5%) tahun 2025, sehingga perlu terus diperhatikan agar tercapainya target tahun 2025. Sebagai catatan, Provinsi Jawa Barat menunjukkan prevalensi wasting terendah sekitar 6%, sementara Provinsi DKI Jakarta mencatat prevalensi wasting tertinggi sebesar 8% pada tahun 2022 di Pulau Jawa. Angka ini menunjukkan pencapaian target SDGs terhadap kondisi wasting masih jauh dari mencapai sasaran. Oleh karena itu, penelitian ini menggunakan Multivariate Adaptive Regression Spline (MARS) yang unggul dalam membentuk fungsi regresi tanpa mengharuskan asumsi terhadap bentuk fungsional tertentu. Hasil penelitian menunjukkan prevalensi balita wasted tertinggi sebesar 16,5% di Kabupaten Pasuruan. Lalu, model MARS yang terbaik adalah 26 fungsi basis, 1 observasi minimum, dan 3 interaksi maksimum. Model ini dapat menjelaskan 92% keragaman variabel dependen dan variabel prediktor yang paling berpengaruh dalam model adalah imunisasi dasar lengkap. Sebaliknya, variabel bayi berjenis kelamin laki-laki tidak digunakan, sehingga tidak berpengaruh sama sekali.
==============================================================================================================================
Malnutrition in toddlers can cause irreversible damage. Wasting, also known as acute malnutrition, occurs due to a lack of quality food intake and is often accompanied by repeated attacks of diseases such as diarrhea, measles, and malaria. Early wasting also increases the risk of stunted growth later in life, including the risk of experiencing wasting and stunting simultaneously, thereby increasing the risk of death. In 2022, the prevalence of wasting in toddlers in Indonesia was 7.7%, which has not reached the Sustainable Development Goals target (5%) in 2025, so it needs to be continuously monitored in order to achieve the 2025 target. For the record, West Java Province showed the lowest wasting prevalence of around 6%, while DKI Jakarta Province recorded the highest wasting prevalence of 8% in 2022 on the island of Java. This figure shows that the achievement of the SDGs target for wasting conditions is still far from reaching the target. Therefore, this study uses Multivariate Adaptive Regression Spline (MARS) which excels in forming regression functions without requiring assumptions about certain functional forms. The results showed the highest prevalence of wasted toddlers at 16.5% in Pasuruan Regency. Then, the best MARS model is 26 basis functions, 1 minimum observation, and 3 maximum interactions. This model can explain 92% of the diversity of the dependent variable and the most influential predictor variable in the model is complete basic immunization. On the other hand, the variable of male babies is not used, so it has no effect at all.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Java Island, Multivariate Adaptive Regression Spline, Toddler, Wasting, Balita, Pulau Jawa
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Ben Hugo
Date Deposited:	09 Aug 2024 03:21
Last Modified:	28 Aug 2024 07:22
URI:	http://repository.its.ac.id/id/eprint/115160

Actions (login required)

View Item