Simanjuntak, Suprianto (2017) Pemodelan Jumlah Kasus HIV dan AIDS di Kota Surabaya Menggunakan Bivariate Generalized Poisson Regression. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1315105003-Undergraduate_Theses.pdf - Published Version Download (8MB) | Preview |
Abstract
HIV dan AIDS merupakan penyakit yang menjadi perhatian di Indonesia, jumlah kasus HIV dan AIDS sejak 1987 sampai 2014 menunjukkan bahwa Provinsi Jawa Timur berada pada urutan ke dua dengan jumlah kasus HIV dan AIDS tertinggi di Indonesia. Kota Surabaya merupakan kota yang memliki kasus baru HIV dan AIDS terbanyak di Provinsi Jawa Timur. Jumlah kasus HIV dan AIDS di Kota Surabaya memiliki korelasi, mengalami kasus overdispersi, serta berdistribusi bivariate generalized Poisson. Dalam rangka mengidentifikasi variabel-variabel prediktor yang berpengaruh signifikan terhadap jumlah kasus HIV dan AIDS, analisis regresi yang diterapkan juga disesuaikan dengan distribusi variabel-variabel respon. Mengaplikasikan bivariate generalized Poisson regression, berdasarkan hasil analisis diperoleh simpulan bahwa faktor yang berpengaruh signifikan tehadap jumlah kasus HIV adalah persentase penduduk pengguna kondom, persentase penduduk yang tamat SMA, persentase penduduk kelompok umur 25-29 tahun dan persentase kegiatan penyuluhan kesehatan masyarakat. Faktor yang berpengaruh signifikan terhadap jumlah kasus AIDS adalah persentase penduduk yang memiliki jaminan kesehatan masyarakat miskin. Model terbaik yang diperoleh adalah model yang melibatkan semua variabel prediktor dengan nilai AIC sebesar 220,038.
=================================================================
HIV and AIDS are the most concernable diseases in Indonesia. In Indonesia, since 1987 to 2014 number of HIV and AIDS case in East Java is ranked at second worst. Surabaya has the highest new number of HIV and AIDS case in East Java. The number of HIV and AIDS case in Surabaya has linear correlations, overdispersion and also has bivariate generalized Poisson distribution. To identify predictor variables which significant to them, regression method must be adjusted with their distribution. According to the results by using bivariate generalized Poisson regression, the significant factors of HIV case number are the percentage of the condom users, the percentage of high school graduated person, the percentage of group aged 25-29 and the percentage of introduction regarding the disease itself. In the other hand, the significant factor regarding AIDS case number is the percentage of civilian that have health guarantee for poor people. The best model that had been acquired is model that is involved all of variable predictor with the number of AIC 220,038.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Sim p |
| Uncontrolled Keywords: | AIDS, bivariate generalized Poisson regression, HIV |
| Subjects: | H Social Sciences > HA Statistics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Suprianto Simanjuntak . |
| Date Deposited: | 14 Dec 2017 09:26 |
| Last Modified: | 05 Mar 2019 07:48 |
| URI: | http://repository.its.ac.id/id/eprint/47871 |
Actions (login required)
 |
View Item |