Prediksi Financial Distress Bank Umum di Indonesia dengan Metode Generalized Extreme Value Regression, Regresi Logistik, dan Analisis Diskriminan Kernel

Widyarani, Adriani (2018) Prediksi Financial Distress Bank Umum di Indonesia dengan Metode Generalized Extreme Value Regression, Regresi Logistik, dan Analisis Diskriminan Kernel. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
1314100106-Undergraduate-Theses.pdf - Published Version
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Stabilitas sistem keuangan telah menjadi perhatian pemerintah dan pihak-pihak terkait. Secara umum, stabilitas sistem keuangan ditopang oleh dua pilar utama yaitu stabilitas harga dan stabilitas sektor keuangan. Perbankan merupakan sektor yang sejauh ini mendominasi sektor keuangan Indonesia. Namun, sektor perbankan cenderung rentan terhadap gangguan. Penelitian ini bertujuan untuk memprediksi kondisi financial distress bank umum di Indonesia dengan metode analisis diskriminan, regresi logistik dan generalized extreme value regression. Pemodelan dilakukan secara univariat, serentak, dan dengan menggunakan seleksi variabel. Regresi logistik mampu menghasilkan nilai AUC tertinggi pada pemodelan yang dilakukan secara univariat dan stepwise. Namun, GEVR mampu menghasilkan AUC yang tinggi dengan lebih banyak variabel yang signifikan apabila pemodelan dilakukan secara serentak. Metode GEVR menghasilkan dua variabel yang signifikan pada window size 1 dan satu variabel signifikan pada window size 3, sedangkan metode regresi logistik tidak menghasilkan variabel yang signifikan apabila pemodelan dilakukan secara serentak. ================ The stability of financial system has been the concern of Indonesian government and related parties. In general, the stability of the financial system is sustained by two main things. They are the stability of Indonesian market and the stability of Indonesian financial sector. In Indonesia, the banking sector is the biggest sector in financial industries. However, the banking sector tends to be risky to disruption. The aim of this research is to predict the financial distress in Indonesian commercial bank using discriminant analysis, logistic regression, and generalized extreme value regression. Modelling is done by using three alternatives, they are univariate modelling, multivariate modelling, and modelling by using variable selection. Logistic regression produces the highest AUC value when univariate modelling or stepwise method is performed. However, GEVR produces high AUC value with more significant variables when the multivariate modelling is performed. The GEVR method produces two significant variables in window size 1 and one variable in window size 3, whereas logistic regression does not produce any significant variables.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Analisis Diskriminan Kernel; Bank Umum; Financial Distress; Generalized Extreme Value Regression; Regresi Logistik; Commercial Bank; Financial Distress Generalized Extreme Value Regression; Kernel Discriminant Analysis, Logistic Regression
Subjects: H Social Sciences > HA Statistics
H Social Sciences > HB Economic Theory
H Social Sciences > HB Economic Theory > Economic forecasting--Mathematical models.
H Social Sciences > HG Finance
Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Mathematics and Science > Statistics > (S1) Undergraduate Theses
Depositing User: Adriani Widyarani
Date Deposited: 09 Apr 2018 03:30
Last Modified: 09 Apr 2018 03:30
URI: http://repository.its.ac.id/id/eprint/50659

Actions (login required)

View Item View Item