Peramalan Jumlah Klaim Produk Asuransi Bumida Menggunakan Metode Autoregressive Integrated Moving Average

Hafizhah, Anindya Zhafira Noer and Syifa, Ayumi Auliyah (2024) Peramalan Jumlah Klaim Produk Asuransi Bumida Menggunakan Metode Autoregressive Integrated Moving Average. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5006211013_5006211094-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penelitian ini dilakukan sebagai bagian dari kerja praktik di Asuransi Syariah Bumida Cabang Surabaya dengan tujuan untuk meramalkan jumlah klaim produk asuransi secara mingguan menggunakan metode Autoregressive Integrated Moving Average (ARIMA). Data yang digunakan merupakan data jumlah klaim per minggu selama periode Januari 2023 hingga Mei 2024. Tahapan analisis meliputi identifikasi stasioneritas data, pemilihan model ARIMA melalui analisis ACF dan PACF, estimasi parameter, uji signifikansi, dan evaluasi diagnostik terhadap residual model. Berdasarkan hasil analisis, diperoleh bahwa model ARIMA(2,0,2) merupakan model terbaik dengan memenuhi asumsi white noise dan residual yang berdistribusi normal, serta memiliki nilai Akaike’s Information Criterion (AIC) paling rendah. Model ini kemudian digunakan untuk melakukan peramalan jumlah klaim selama sepuluh minggu ke depan. Hasil peramalan ini diharapkan dapat menjadi referensi bagi manajemen perusahaan dalam mengelola risiko klaim dan perencanaan keuangan yang lebih efektif.
=====================================================================================================================================
This study was conducted as part of an internship program at Asuransi Syariah Bumida Surabaya Branch, aiming to forecast the weekly number of insurance claims using the Autoregressive Integrated Moving Average (ARIMA) method. The data utilized consists of weekly claim numbers from January 2023 to May 2024. The analysis stages include stationarity identification, ARIMA model selection through ACF and PACF analysis, parameter estimation, significance testing, and diagnostic evaluation of model residuals. The results indicate that the ARIMA(2,0,2) model is the best-fit model as it satisfies the white noise assumption, has normally distributed residuals, and exhibits the lowest Akaike’s Information Criterion (AIC) value. This model is then employed to forecast insurance claims for the subsequent ten weeks. The forecasting results are expected to serve as a valuable reference for company management in managing claim risks and planning financial strategies more effectively.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	ARIMA, Asuransi, Klaim Produk, Peramalan, Runtun Waktu, ARIMA,Forecasting, Product Claims, Time Series, Insurance
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Anindya Zhafira Noer Hafizhah
Date Deposited:	18 Jul 2025 05:56
Last Modified:	18 Jul 2025 05:56
URI:	http://repository.its.ac.id/id/eprint/120038

Actions (login required)

View Item