| dc.contributor.advisor | Budiarti, Retno | |
| dc.contributor.advisor | Purnaba, I Gusti Putu | |
| dc.contributor.author | VELANI, ARISKA CITRA | |
| dc.date.accessioned | 2026-06-19T06:45:36Z | |
| dc.date.available | 2026-06-19T06:45:36Z | |
| dc.date.issued | 2026 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/173534 | |
| dc.description.abstract | Perusahaan asuransi perlu menyiapkan cadangan klaim untuk memenuhi kewajiban pembayaran klaim di masa mendatang secara tepat. Penelitian ini bertujuan untuk memprediksi cadangan klaim reported but not settled (RBNS) menggunakan pendekatan generalized linear model (GLM) dengan sebaran lognormal serta mengevaluasi tingkat akurasi model menggunakan mean absolute percentage error (MAPE). Data yang digunakan berupa data klaim kumulatif produk homeowners’ yang disusun dalam bentuk run-off triangle berukuran 6×6 dan kemudian ditransformasikan menjadi data incremental. Hasil analisis menunjukkan bahwa data memiliki nilai mean sebesar 4,165,833 dan varians sebesar 1.69869× 10^15, yang sesuai dengan karakteristik sebaran lognormal. Estimasi menggunakan metode GLM menghasilkan total cadangan klaim sebesar 2,091,730 dengan nilai MAPE sebesar 4.95%, yang menunjukkan bahwa model memiliki tingkat akurasi yang sangat baik. Hasil penelitian ini menunjukkan bahwa metode GLM dengan sebaran lognormal layak digunakan untuk mengestimasi cadangan klaim reported but not settled (RBNS). | |
| dc.description.abstract | Insurance companies are required to establish adequate claim reserves to fulfill future claim payment obligations accurately. This study aims to estimate reported but not settled (RBNS) claim reserves using a generalized linear model (GLM) with a lognormal distribution and to evaluate the model’s predictive accurancy using the mean absolute percentage error (MAPE). The study uses cumulative claim data from homeowners’ insurance products arranged in a 6×6 run-off triangle, which are subsequently transformed into incremental claim data. The results indicate that the data have a mean of 4,165,833 and a variance of 1.69869× 10^15, suggesting that the lognormal distribution is appropriate for modeling the claim amounts. The GLM approach produces a total claim reserve of 2,091,730 with a MAPE value of 4.95%, indicating a high level of predictive accuracy. These findings demonstrate that the GLM with a lognormal distribution provides an appropriate and reliable approach for estimating RBNS claim reserves. | |
| dc.description.sponsorship | Orang tua | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Estimasi Cadangan Klaim Reported but Not Settled menggunakan pendekatan Generalized Linear Model dengan Sebaran Lognormal | id |
| dc.title.alternative | Estimation of Reported but Not Settled Claim Reserves Using a Generalized Linear Model with a Lognormal Distribution | |
| dc.type | Skripsi | |
| dc.subject.keyword | claim reserves | id |
| dc.subject.keyword | distribution | id |
| dc.subject.keyword | generalized linear model | id |
| dc.subject.keyword | mean absolute percentage error | id |
| dc.subject.keyword | reported but not settled | id |