Andriani, Levi Tri (2019) Menentukan model terbaik menggunakan kriteria Informasi ϕ-divergensi pada model Log Linier Multivariat. Diploma thesis, UIN Sunan Gunung Djati Bandung.
|
Text (COVER)
1_cover.pdf Download (313kB) | Preview |
|
|
Text (ABSTRAK)
2_abstrak.pdf Download (403kB) | Preview |
|
|
Text (DAFTAR ISI)
3_daftarisi.pdf Download (459kB) | Preview |
|
|
Text (BAB I)
4_bab1.pdf Download (529kB) | Preview |
|
![[img]](https://digilib.uinsgd.ac.id/style/images/fileicons/text.png) |
Text (BAB II)
5_bab2.pdf Restricted to Registered users only Download (521kB) | Request a copy |
|
|
Text (BAB III)
6_bab3.pdf Restricted to Registered users only Download (549kB) | Request a copy |
|
|
Text (BAB IV)
7_bab4.pdf Restricted to Registered users only Download (371kB) | Request a copy |
|
|
Text (BAB V)
8_bab5.pdf Restricted to Registered users only Download (307kB) | Request a copy |
|
|
Text (DAFTAR PUSTAKA)
9_daftarpustaka.pdf Restricted to Registered users only Download (397kB) | Request a copy |
Abstract
INDONESIA: Pemilihan model dapat dilakukan melalui kriteria informasi seperti Akaike Information Criterian (AIC) dan Bayesian Information Criterian (BIC). Namun kriteria informasi tersebut hanya dapat digunakan untuk sampel yang tidak terlalu besar karena bentuknya yang sederhana. Skripsi ini membahas tentang pendekatan kriteria informasi ϕ-divergensi untuk kondisi sampel yang besar. Model yang digunakan dalam skripsi ini adalah model log linier multivariat empat dimensi yaitu model yang memiliki empat variabel kategori. Estimasi parameter model menggunakan metode Maximum Likelihood Estimation (MLE) dengan uji kecocokan model menggunakan likelihood ratio square. Penerapan model diimplementasikan pada data tentang pendapatan asli daerah berdasarkan jenis daerah, jumlah penduduk dan luas wilayah di Indonesia pada tahun 2018 dengan 508 kabupaten dan kota. Dari studi kasus tersebut terdapat 23 kombinasi model dan diperoleh 5 model yang memenuhi uji kecocokan model yang selanjutnya ditentukan model yang terbaik dengan menggunakan kriteria informasi ϕ-divergensi. ENGLIS: Selection of model can initially use information criteria such as Akaike Information Criterian (AIC) and Baesian Information Criterian (BIC). But the information criteria can only be used for samples that are not too large because of their simple form. This thesis discusses the ϕ-divergence information criterion approach for large sample conditions. The model used in this thesis is a four-dimensional multivariate log linear model that is a model that has four categorical variables. Estimation of model parameters using the Maximum Likelihood Estimation (MLE) method with the model compatibility test using the likelihood ratio square. The application on of the model was implemented on data on local own revenue based on the type of region, population and area in Indonesia in 2018 with 508 districts and cities. From the case study there were 23 model combinations and 5 models were obtained that met the model fit test which was then determined by the best model using the ϕ-divergence information criteria
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Data Kategori; Model Log Linier Multivariat Empat Dimensi; MLE; Pemilihan Model; Kriteria Informasi ϕ-divergensi |
| Subjects: | Mathematics > Data Processing and Analysis of Mathematics |
| Divisions: | Fakultas Sains dan Teknologi > Program Studi Matematika |
| Depositing User: | Levi Tri Andriani |
| Date Deposited: | 27 Sep 2019 03:14 |
| Last Modified: | 27 Sep 2019 03:14 |
| URI: | https://digilib.uinsgd.ac.id/id/eprint/24065 |
Actions (login required)
 |
View Item |
