Pemodelan Kasus HIV/AIDS Di Provinsi Jawa Barat Menggunakan Spatial Autoregressive Moving Average dan Spatial Durbin Model

Putri, Khalida Aurora Amanda (2025) Pemodelan Kasus HIV/AIDS Di Provinsi Jawa Barat Menggunakan Spatial Autoregressive Moving Average dan Spatial Durbin Model. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211022-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Tingginya disparitas kasus HIV antarwilayah di Jawa Barat menjadi tantangan dalam upaya pengendalian yang efektif. Kabupaten/kota dengan kasus tinggi umumnya berdekatan dengan pusat urbanisasi dan mobilitas tinggi, sedangkan wilayah lain menunjukkan angka yang jauh lebih rendah. Ketimpangan ini menunjukkan potensi adanya keterkaitan spasial yang belum banyak dijelaskan dalam pendekatan regresi linear berganda. Penelitian ini bertujuan untuk memodelkan log kasus HIV antarwilayah di Provinsi Jawa Barat dengan mempertimbangkan faktor spasial dan sosial sebagai dasar kebijakan kesehatan berbasis wilayah. Metode yang digunakan mencakup Spatial Autoregressive Moving Average (SARMA), Spatial Autoregressive (SAR), Spatial Error Model (SEM), Spatial Durbin Model (SDM), dan Spatial Durbin Error Model (SDEM) dengan pendekatan pembobot spasial queen contiguity. Setelah melalui seleksi variabel dengan metode stepwise, diperoleh tiga prediktor utama yaitu rasio KDRT, rasio peserta KB kondom pria, dan rasio kelompok PIK-R. Hasil uji Moran’s I pada kasus HIV sebesar 0,528 menunjukkan adanya dependensi spasial positif yang signifikan antarwilayah. Model SDEM terpilih sebagai model terbaik dengan AIC sebesar 48,815. Model memberikan nilai R-squared sebesar 69,72% sehingga mampu menangkap pengaruh tidak langsung dari wilayah sekitar melalui lag variabel prediktor, serta memberikan gambaran distribusi spasial yang relevan bagi analisis lanjutan. Temuan ini menegaskan pentingnya pendekatan spasial, khususnya model SDEM, dalam memahami karakteristik penyebaran HIV/AIDS di Jawa Barat.
==================================================================================================================================
The significant disparity in HIV cases across regions in West Java presents a challenge for effective disease control. Districts and cities with high case counts are generally clustered around centers of urbanization and high mobility, while other areas report considerably lower numbers. This disparity suggests the presence of spatial interdependence that is not adequately captured by traditional multiple linear regression approaches. This study aims to model the natural logarithm of HIV cases across regions in West Java Province by incorporating spatial and social factors as a basis for region-based public health analysis. The methods employed include Spatial Autoregressive Moving Average (SARMA), Spatial Autoregressive (SAR), Spatial Error Model (SEM), Spatial Durbin Model (SDM), and Spatial Durbin Error Model (SDEM), using a queen contiguity spatial weight matrix. Following variable selection via stepwise regression, three key predictors were identified: the ratio of domestic violence (KDRT) cases, the ratio of male condom contraceptive users, and the ratio of PIK-R (Youth Information and Counseling Center) groups. The Moran’s I test for HIV cases yielded a value of 0.528, indicating significant positive spatial dependence between regions. The SDEM model was selected as the best-fitting model, with an AIC value of 48.815. This model achieved an R-squared of 69.72%, effectively capturing indirect spatial effects from neighboring regions through lagged predictor variables, while providing a spatial distribution overview relevant for further analysis. These findings highlight the importance of spatial approaches, particularly the SDEM model, in understanding the characteristics of HIV/AIDS transmission patterns across West Java.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	HIV, Regresi Spasial, SDEM, Jawa Barat, Spatial Regression, West Java
Subjects:	H Social Sciences > HA Statistics > HA30.6 Spatial analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Khalida Aurora Amanda
Date Deposited:	23 Jul 2025 07:13
Last Modified:	23 Jul 2025 07:13
URI:	http://repository.its.ac.id/id/eprint/120806

Actions (login required)

View Item