Penerapan Generalized Extreme Value Regression Additive Models Dalam Mendeteksi Potensi Financial Distress Pada Perusahaan Asuransi Jiwa di Indonesia

Mutmainna, Rifdah (2025) Penerapan Generalized Extreme Value Regression Additive Models Dalam Mendeteksi Potensi Financial Distress Pada Perusahaan Asuransi Jiwa di Indonesia. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211089-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (8MB) | Request a copy

Abstract

Perusahaan asuransi jiwa memegang peran penting dalam perekonomian dengan mengelola dana premi nasabah sebagai tabungan nasional untuk pembangunan dan investasi. Namun, terdapat perusahaan asuransi jiwa di Indonesia menghadapi kondisi financial distress, yang dapat mengarah pada kebangkrutan. Penelitian ini bertujuan memprediksi financial distress pada perusahaan asuransi jiwa menggunakan metode Generalized Extreme Value Regression (GEVR) Additive Model. Metode ini mengacu pada kejadian extreme imbalance proportion dimana perusahaan dengan kondisi financial distress memiliki proporsi yang lebih sedikit dibandingkan perusahaan sehat dan dilengkapi smoothing parameter untuk menangani hubungan nonlinear antar variabel dan mengurangi risiko overfitting. Indikator financial distress ditentukan oleh Risk Based Capital (RBC) < 120%. Analisis dilakukan menggunakan time window size dengan pendekatan drift serta seleksi variabel dengan feature selection pada variabel yang terindikasi masalah multikolinearitas dan eliminasi backward. Hasil menunjukkan bahwa model terbaik untuk time window size 0 dan 2 dalam memprediksi financial distress adalah GEVR dengan eliminasi backward, serta GEVR serentak untuk size 1 dan 3. Rasio keuangan seperti CER, ITRR, dan TRR konsisten signifikan pada mayoritas metode dan size. Selain itu, indikator makroekonomi seperti inflasi dan BI7DRR (suku bunga) juga memengaruhi kondisi financial distress pada beberapa size. penelitian ini menekankan pentingnya pengelolaan rasio keuangan yang signifikan serta pemodelan risiko yang adaptif untuk menjaga stabilitas keuangan perusahaan asuransi jiwa di Indonesia. Secara umum, semua metode dalam penelitian memberikan hasil terbaik pada size 0. Hal ini dapat disebabkan karena penggunaan rasio keuangan saat ini akan lebih relevan untuk memprediksi kondisi perusahaan asuransi jiwa pada tahun yang sama.
===================================================================================================================================
Life insurance companies play a crucial role in the economy by managing customer premium funds as national savings for development and investment. However, some life insurance companies in Indonesia face financial distress, which can lead to bankruptcy. This study aims to predict financial distress in life insurance companies using the Generalized Extreme Value Regression (GEVR) Additive Model. This method refers to the extreme imbalance proportion event where companies with financial distress have a smaller proportion than healthy companies and features smoothing parameters to handle nonlinear relationships between variables and reduce the risk of overfitting. Financial distress is indicated by a Risk Based Capital (RBC) of less than 120%. The analysis employs a time window size with a drift approach and variable selection through feature selection to address multicollinearity issues and backward elimination. The results indicate that the best model for predicting financial distress is GEVR with backward elimination for time window sizes 0 and 2, while the concurrent GEVR model performs best for sizes 1 and 3. Financial ratios such as CER, ITRR, and TRR are consistently significant in most methods and sizes. Additionally, macroeconomic indicators like inflation and BI7DRR (interest rate) influence financial distress in certain time window sizes. This study underscores the importance of managing significant financial ratios and employing adaptive risk modeling to ensure the financial stability of life insurance companies in Indonesia. In general, all methods in the study provide the best results at size 0. This can be caused because the use of current financial ratios will be more relevant to predicting the condition of life insurance companies in the same year.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Life Insurance, Financial Distress, GEV Regression, GAM, Smoothing Parameter, Asuransi Jiwa, GEVR Additive
Subjects:	H Social Sciences > HC Economic History and Conditions H Social Sciences > HC Economic History and Conditions > HC441 Macroeconomics. H Social Sciences > HG Finance H Social Sciences > HG Finance > HG8051 Insurance H Social Sciences > HG Finance > HG8771 Life insurance
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Rifdah Mutmainna
Date Deposited:	26 Jan 2025 10:33
Last Modified:	26 Jan 2025 10:33
URI:	http://repository.its.ac.id/id/eprint/116892

Actions (login required)

View Item