Penentuan Keputusan dan Strategi Optimal untuk Meminimasi Risiko Kredit Macet dengan Metode Data Mining dan Inspection Game

Rochmadhan, Oryza Akbar (2017) Penentuan Keputusan dan Strategi Optimal untuk Meminimasi Risiko Kredit Macet dengan Metode Data Mining dan Inspection Game. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
2513100057_Undergraduate Thesis.pdf - Published Version

Download (2MB) | Preview

Abstract

Kredit macet merupakan salah satu masalah besar yang paling sering dihadapi oleh Bank. Kerugian yang dialami bank dikarenakan permasalahan ini terhitung besar. Keadaan yang dihadapi sekarang, proporsi jumlah kredit lancar berada pada tingkat 83,99%. Pihak Bank berupaya untuk menaikkan proporsi jumlah kredit lancar hingga 95% dengan cara menekan jumlah kredit macet. Dalam penelitian ini, digunakan penerapan data mining dan game theory dalam menyusun model decision tree dan inspection game untuk dapat menghasilkan suatu kebijakan yang dapat menekan jumlah kredit macet. Model decision tree digunakan untuk mendapat metode prediksi status nasabah selama peminjaman, sedangkan model inspection game digunakan untuk mendapatkan kebijakan yang dapat menurunkan tendensi nasabah untuk melakukan kredit macet. Hasil dari penelitian adalah, tingkat akurasi prediksi dari model decision tree: nilai POD sebesar 99,50%, TNR sebesar 100%, dan OA sebesar 99,80%. Model inspection game merumuskan kebijakan yang dapat menurunkan tendensi nasabah untuk melakukan kredit macet adalah peningkatan suku bunga dan lama waktu tenor. ========================================================================================================= Bad debt is one of many big problems which is frequently experienced by Bank X. The loss which is caused by bad debt is considered big. The existing condition, is only 83,99% debt that given by bank is considered as a good debt. Bank X is trying to raise the proportion to 95% by suppressing the bad debt. This research will use data mining and game theory as methods for constructing decision tree model and inspection game that will results in having policy/ies that will suppress bad debt proportion.The decision tree model is used to get prediction method of customers’s states during the tenor period, while inspection game model is used to build policy/ies that will reduce the likeliness of customers doing bad debt. The results of this research are, the accuracy of prediction from decision tree model: the value of POD, TNR, and OA are 99,50%, 100%, and 99,80% respectively, the policy/ies from inspection game model that will reduce the likeliness of customers to do bad debt are increasing the loan rate and the tenor period.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Decision Tree, Inspection Game, Kebijakan, Kredit Macet
Subjects: H Social Sciences > HG Finance
Q Science > QA Mathematics > QA269 Game theory
Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis
Divisions: Faculty of Industrial Technology > Industrial Engineering > (S1) Undergraduate Theses
Depositing User: Oryza Akbar Rochmadhan
Date Deposited: 20 Sep 2017 02:01
Last Modified: 01 Mar 2019 07:37
URI: http://repository.its.ac.id/id/eprint/43642

Actions (login required)

View Item View Item