Rulitasari, Diana Desi (2016) Pemodelan Prevalensi Penyakit Jantung Koroner Dengan Pendekatan Geographically Weighted Regression Di Jawa Timur Tahun 2013. Undergraduate thesis, Institut Technology Sepuluh Nopember.
Preview |
Text
1314105054-Undergraduate Thesis.pdf - Published Version Download (3MB) | Preview |
Abstract
Penyakit jantung koroner (PJK) adalah penyempitan pembuluh darah
kecil yang memasok darah dari oksigen ke jantung. Ini disebabkan oleh
pembentukan plak pada dinding arteri. Berdasarkan diagnosis dan
gejala, estimasi jumlah penderita PJK terbanyak terdapat di Jawa
Timur sebanyak 375.127 orang. Variabel yang memengaruhi PJK
diantaranya obesitas, hipertensi, kebiasaan merokok, aktivitas fisik, dan
ekonomi. Beberapa penelitian tentang PJK, sebagian besar tidak
menekankan aspek humaniora seperti kekhasan budaya yang
direpresentasikan kekhasan lokasi (kabupaten/kota). Oleh karena itu
penelitian ini dikembangkan pemodelan PJK yang mengakomodasi
adanya aspek perilaku masyarakat yang direpresentasikan dalam
spasial (lokasi). Metode yang digunakan dalam penelitian ini dengan
pendekatan Geographically Weighted Regression (GWR). GWR
merupakan model regresi untuk variabel respon yang bersifat kontinu
dimana pemodelan memperhatikan letak geografis suatu wilayah.
Penelitian ini bertujuan memodelkan prevalensi PJK berdasarkan
variabel yang mempengaruhinya. Berdasarkan hasil analisis terbentuk
dua kelompok persebaran kab/kota berdasarkan variabel yang
signifikan. Pada kelompok pertama variabel persentase penderita
diabetes melitus dan persentase merokok tiap hari berpengaruh
terhadap prevalensi PJK. Pada kelompok kedua variabel persentase
penderita diabetes melitus, merokok tiap hari, dan aktivitas fisik kurang
aktif berpengaruh terhadap prevalensi PJK. Berdasarkan kriteria nilai
AIC dan R2 diperoleh bahwa model GWR yang lebih baik, karena
mempunyai nilai AIC terkecil yaitu 71,043 dan R2 tertinggi yaitu 54,1%.
=============================================================================================================
Coronary heart disease (CHD) is a narrowing of the small blood vessels
that supply oxygen to the blood from the heart. This is caused by the
formation of plaque on artery walls. Based on the diagnosis and
symptoms, the estimated number of people with coronary heart disease
are highest in East Java province as much as 375 127. Variables that
influence CHD including obesity, hypertension, smoking, physical
activity, and economic. From some research on CHD, most do not
emphasize this aspect of the humanities such as cultural peculiarities
which represented the peculiarities of the location (district/city).
Therefore in this study developed CAD modeling that accommodates their
behavioral aspects of society are represented in the spatial (location). The
method used in this study with the approach Geographically Weighted
Regression (GWR). GWR is a regression model for the response variable
is continuous where modeling attention to the geographical location of a
region. This study aims to model the prevalence of CHD based on
variables that influence. Based on the analysis formed two groups. In the
first group variable percentage of patients with diabetes mellitus and
smoking every day percentage effect on the prevalence of CHD. In the
second group variable percentage of patients with diabetes mellitus,
smoking every day, and physical activity were less active influence on the
prevalence of CHD. Based on the criteria AIC and R2 values obtained
that GWR model is better, because it has the smallest AIC value is 71.043
and the highest R2 is 54.1%.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Rul p |
| Uncontrolled Keywords: | AIC, GWR, PJK, R2 |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Mr. Tondo Indra Nyata |
| Date Deposited: | 03 Jan 2020 08:57 |
| Last Modified: | 03 Jan 2020 08:57 |
| URI: | http://repository.its.ac.id/id/eprint/72531 |
Actions (login required)
 |
View Item |