Sitompul, Madaleni Ranti (2022) Analisis data longitudinal dengan menggunakan Weighted Generalized Estimating Equations (WGEE) pada data hilang. Sarjana thesis, UIN Sunan Gunung Djati Bandung.
|
Text (COVER)
1_cover.pdf Download (203kB) | Preview |
|
|
Text (ABSTRAK)
2_abstrak.pdf Download (282kB) | Preview |
|
|
Text (DAFTAR ISI)
3_daftarisi.pdf Download (460kB) | Preview |
|
|
Text (BAB I)
4_bab1.pdf Download (400kB) | Preview |
|
Text (BAB II)
5_bab2.pdf Restricted to Registered users only Download (877kB) | Request a copy |
||
Text (BAB III)
6_bab3.pdf Restricted to Registered users only Download (788kB) | Request a copy |
||
Text (BAB IV)
7_bab4.pdf Restricted to Registered users only Download (573kB) | Request a copy |
||
Text (BAB V)
8_bab5.pdf Restricted to Registered users only Download (376kB) | Request a copy |
||
Text (DAFTAR PUSTAKA)
9_daftarpustaka.pdf Restricted to Registered users only Download (441kB) | Request a copy |
Abstract
INDONESIA : Data longitudinal merupakan data hasil pengukuran dari waktu ke waktu pada satu atau beberapa variabel di setiap individu/objek yang sama. Analisis data longitudinal, terkendala ketika terjadi data hilang atau data yang tidak lengkap. Salah satu analisis data longitudinal adalah dengan menggunakan metode GEE, tetapi metode akan bias ketika adanya data hilang. Penelitian ini membahas alternatif metode ketika terjadi data hilang yaitu metode Weighted Generalized Estimating Equations (WGEE) yang memberikan bobot yang baik sehingga kemudian model yang dihasilkan dapat diestimasi dengan hasil yang baik dan tak bias. Implementasi metode WGEE diterapkan pada data Indeks Pembangunan Kesehatan Masyarakat (IPKM) Kabupaten / Kota di Sumatera Utara dengan 7 variabel independen yaitu : kesehatan balita, kesehatan reproduksi, pelayanan kesehatan, perilaku kesehatan, penyakit tidak menular, penyakit menular, dan kesehatan lingkungan. Menghasilkan bobot yang berbeda beda setiap observasinya, hasil analisis diperelah terdapat 2 variabel independen yang tidak mempengaruhi IPKM dengan tingkat signifikansi 95% yaitu penyakit menular dan kesehatan balita, sedangkan variabel lain berpengaruh secara signifikan. ENGLISH : Longitudinal data is data measured from time to time on one or several variables in each of the same individual/object. Longitudinal data analysis is constrained when missing data or incomplete data occur. One of the longitudinal data analysis is to use the GEE method, but the method will be biased when there is missing data. This study discusses an alternative method when missing data occurs, namely the Weighted Generalized Estimating Equations (WGEE) method which provides good weight so that the resulting model can be estimated with good and unbiased results. The implementation of the WGEE method was applied to District/City Community Health Development Index (IPKM) data in North Sumatera with 7 independent variables namely: toddler health, reproductive health, health services, health behavior, non-communicable diseases, infectious diseases, and environmental health. Resulting in different weights for each observation, the results of the analysis found that there were 2 independent variables that did not affect IPKM with a 95% significance level, namely infectious diseases and toddler health, while other variables had a significant effect.
Item Type: | Thesis (Sarjana) |
---|---|
Uncontrolled Keywords: | Longitudinal data; Missing data; IPKM; WGEE |
Subjects: | Mathematics Mathematics > Data Processing and Analysis of Mathematics Mathematics > Research Methods of Mathematics Applied mathematics > Statistical Mathematics |
Divisions: | Fakultas Sains dan Teknologi > Program Studi Matematika |
Depositing User: | Madaleni Ranti Sitompul |
Date Deposited: | 10 Jan 2023 00:47 |
Last Modified: | 10 Jan 2023 00:47 |
URI: | https://digilib.uinsgd.ac.id/id/eprint/63368 |
Actions (login required)
View Item |