Aplikasi quantile regression untuk perhitungan premi asuransi kendaraan bermotor menggunakan quantile premium principle = Application of quantile regression for the calculation of vehicle insurance premium using quantile premium principle

Sutardiman, Mario (2024) Aplikasi quantile regression untuk perhitungan premi asuransi kendaraan bermotor menggunakan quantile premium principle = Application of quantile regression for the calculation of vehicle insurance premium using quantile premium principle. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (250kB)
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (585kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (591kB)
	Text (Chapter1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (592kB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (712kB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (991kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (580kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (561kB)
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (5MB)

Abstract

Penelitian ini berfokus pada aplikasi Quantile Regression (QR) untuk perhitungan premi asuransi kendaraan bermotor menggunakan Quantile Premium Principle (QPP). Dalam industri asuransi, perhitungan premi yang akurat dan adil merupakan hal krusial untuk mengelola risiko finansial. Penelitian ini membandingkan efektivitas QPP dengan Expected Value Premium Principle (EVPP) dalam konteks perhitungan premi asuransi. Model yang digunakan meliputi Zero-Inflated Poisson (ZIP) untuk frekuensi klaim, Inverse Gaussian Generalized Linear Model (GLM) untuk besar klaim, dan QR untuk kuantil besar klaim. Pada penelitian ini, variabel prediktor yang digunakan untuk setiap model adalah DriverAge dan CarAge yang telah dikategorikan menjadi beberapa kelompok, sehingga memungkinkan perhitungan premi yang bervariasi dari berbagai kelas risiko yang terbentuk berdasarkan kedua variabel tersebut. Hasil penelitian menunjukkan bahwa QR dapat memperkirakan kuantil besar klaim yang selanjutnya digunakan pada perhitungan premi menggunakan QPP. Lebih lanjut lagi, hasil perhitungan premi dengan EVPP dan QPP tidak selalu bergerak ke arah yang sama, karena ukuran risiko yang digunakan kedua prinsip perhitungan premi tersebut berbeda. Ditemukan bahwa QPP lebih efektif dalam hal mengidentifikasi dan mengakomodasi risiko yang beragam dibandingkan dengan EVPP, sehingga dapat menawarkan evaluasi risiko yang lebih komprehensif dalam proses underwriting. Penelitian ini juga menganalisis pengaruh usia pengemudi dan usia kendaraan terhadap variabilitas premi asuransi, menunjukkan dampak signifikan dari kedua faktor tersebut pada perhitungan premi. / This study focuses on the application of Quantile Regression (QR) for calculating vehicle insurance premiums using the Quantile Premium Principle (QPP). In the insurance industry, accurate and fair premium calculation is crucial for managing financial risks. This study compares the effectiveness of QPP with the Expected Value Premium Principle (EVPP) in the context of insurance premium calculation. The models used include Zero-Inflated Poisson (ZIP) for claim frequency, Inverse Gaussian Generalized Linear Model (GLM) for claim severity, and QR for claim severity quantiles. In this study, the predictor variables used for each model are DriverAge and CarAge, which have been categorized into several groups, thereby allowing for the calculation of varied insurance premiums from the different risk classes formed based on these two variables. The results show that QR can estimate quantiles of claim severity, which are then used in premium calculations using QPP. Furthermore, the results of premium calculations using EVPP and QPP do not always move in the same direction, due to the different risk measures employed by these two premium calculation principles. It was found that QPP is more effective in identifying and accommodating diverse risks compared to EVPP, thus offering a more comprehensive risk evaluation in the underwriting process. This study also analyzes the impact of driver age and car age on insurance premium variability, demonstrating the significant influence of these two factors on premium calculation.

Actions (login required)

View Item