Setianingsih, Ni Putu Budi (2014) Pemodelan faktor-faktor yang mempengaruhi penggolongan kredit di PT.Bank X (Persero) Tbk dengan menggunakan metode hybrid genetic algorithm- logistic regression. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1310100047-Undergraduate_Thesis.pdf Download (852kB) | Preview |
Preview |
Text
1310100047-Paper.pdf Download (1MB) | Preview |
Preview |
Text
1310100047-Presentation.pdf Download (23MB) | Preview |
Abstract
PT Bank X (Persero) Tbk menunjukkan kinerja baik dalam
perkreditan sampai pada tahun 2010. Namun sampai pada akhir
kuartal III tahun 2013, PT Bank X (Persero) Tbk menjadi salah
satu bank persero di Indonesia yang mengalami peningkatan
rasio kredit bermasalah atau non performing loan (NPL). Terjadinya
kredit bermasalah akan memberi dampak bagi kreditur
maupun debitur. Untuk itu, penelitian ini melakukan pemodelan
faktor-faktor yang mempengaruhi penggolongan kredit di PT
Bank X (Persero) Tbk guna memprediksi risiko kredit dari calon
debitur. Pemodelan tersebut dilakukan dengan menggunakan
metode regresi logistik dan hybrid genetic algorithm – logistic
regression terhadap data debitur di PT Bank X (Persero) Tbk.
Fungsi fitness yang digunakan adalah ukuran kesesuaian model
regresi logistik, yaitu pseudo R2 atau
dan MSE. Metode hybrid
genetic algorithm – logistic regression memberikan hasil yang
lebih baik karena
yang bernilai lebih tinggi dan MSE yang
bernilai lebih rendah dibandingkan dengan hasil estimasi
parameter model regresi logistik menggunakan metode maximum
likelihood estimation (MLE).
================================================================================================================
PT Bank X (Persero) Tbk performed good profile in credit
issue until 2010. But at the end of third quarter in 2013, PT Bank
X (Persero) Tbk became one of some banks that having increment
in non performing loan (NPL) ratio. NPL will definitely give
affects for both creditors and debtors. Based on that issue, this
research applied two methods for modelling factors that affect
credit classification in PT Bank X (Persero) Tbk: logistic
regression and hybrid genetic algorithm – logistic regression.
The models were then used to predict debtor’s credit risk. Hybrid
genetic algorithm – logistic regression shows better result, since
hybrid genetic algorithm gives higher
and lower MSE than
and MSE from the logistic regression model.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Set p |
| Uncontrolled Keywords: | Algoritma genetika; MSE; penggolongan kredit; pseudo R2; regresi logistik |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | - Taufiq Rahmanu |
| Date Deposited: | 04 Jul 2019 07:06 |
| Last Modified: | 04 Jul 2019 07:06 |
| URI: | http://repository.its.ac.id/id/eprint/63489 |
Actions (login required)
 |
View Item |