Klasifikasi Nasabah Pembiayaan Bermasalah Di PT. BPRS Lantabur Tebuireng Menggunakan Regresi Logistik-Synthetic Minority Over-sampling Technique

Islahulhaq, Islahulhaq (2021) Klasifikasi Nasabah Pembiayaan Bermasalah Di PT. BPRS Lantabur Tebuireng Menggunakan Regresi Logistik-Synthetic Minority Over-sampling Technique. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 10611710000100-Undergraduate_Thesis for D4 Program.pdf] Text
10611710000100-Undergraduate_Thesis for D4 Program.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (1MB) | Request a copy
[thumbnail of 10611710000100-Undergraduate_Thesis.pdf] Text
10611710000100-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy
[thumbnail of 10611710000100-Undergraduate_Thesis.pdf] Text
10611710000100-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Analisa pembiayaan merupakan proses menganalisa kemampuan nasabah bank dalam membayar angsuran dengan tujuan untuk meminimalisir risiko nasabah tidak membayar angsuran atau Non-Performing Financing (NPF). Pada Tahun 2020 rasio NPF mengalami peningkatan, akibat dari turunnya pendapatan masyarakat selama pandemi Covid-19. Fenomena ini menimbulkan ketidaksehatan kinerja perbankan, salah satunya adalah PT BPRS Lantabur Tebuireng. Pada bulan Desember tahun 2020 terdapat 17% nasabah pembiayaan bermasalah di PT BPRS Lantabur Tebuireng. Adanya ketidakseimbangan antara jumlah nasabah pembiayaan lancar dan pembiayaan bermasalah mengakibatkan hasil ketepatan klasifikasi kurang baik. Oleh karena itu, penelitian ini mengklasifikasikan nasabah pembiayaan bermasalah menggunakan metode Regresi Logistik dan Regresi Logistik-Synthetic Minority Over-Sampling Technique (SMOTE). Hasil penelitian ini menunjukkan bahwa model Regresi Logistik-SMOTE adalah model terbaik untuk klasifikasi nasabah pembiayaan bermasalah. Variabel yang berpengaruh signifikan terhadap pembiayaan bermasalah yaitu jangka waktu, jenis penggunaan, jaminan dan jenis pekerjaan. Model Regresi Logistik-SMOTE mampu menangani ketidakseimbangan jumlah data dan meningkatkan nilai spesificity saat menggunakan metode Regresi Logistik dari 0,04 menjadi 0,21, dengan accuracy 0,81, sensitivity 0,94.
================================================================================================
Financing analysis is a process of analysing the ability of bank customers to pay instalments with the aim of minimising the risk of a customer not paying instalments, which is also called Non-Performing Financing (NPF). In 2020 the NPF ratio increased, due to the decline in peoples's income during the Covid-19 pandemic. This phenomenon has led to bad banking performance, one of which is PT BPRS Lantabur Tebuireng. During December 2020 the percentage of Non-Performing Financing customers in the bank was 17%. The imbalance between the number of good-financing and non-performing financing customers has resulted in poor classification accuracy results. Therefore, this study classifies non-performing financing customers using the Logistic Regression and Logistic Regression-Synthetic Minority Over-Sampling Technique(SMOTE) method. The results of this study indicate that the Logistics Regression-SMOTE model is the best model for the classification of non-performing financing customers. The variables that have a significant effect are loan term, type of use, collateral, and type of work. The Logistic Regression-SMOTE can handle the data imbalance and increase the specificity value when using Logistic Regression method form 0.04 to 0.21, with accuracy 0.81, sensitivity 0.94.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: klasifikasi, non-performing financing, regresi logistik, SMOTE, classification, logistic regression, non-performing financing, SMOTE
Subjects: H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics
H Social Sciences > HG Finance
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions: Faculty of Vocational > 49501-Business Statistics
Depositing User: Islahulhaq Islahulhaq
Date Deposited: 15 Aug 2021 22:20
Last Modified: 15 Aug 2021 22:20
URI: http://repository.its.ac.id/id/eprint/86889

Actions (login required)

View Item View Item