Perbandingan Metode Exponential Smoothing Dan Autoregressive Integrated Moving Average Dalam Peramalan Jumlah Realisasi Kredit Tahun 2024

Silaban, Yola Jovita and Mahendra, Adi (2024) Perbandingan Metode Exponential Smoothing Dan Autoregressive Integrated Moving Average Dalam Peramalan Jumlah Realisasi Kredit Tahun 2024. Project Report. [s.n], [s.l.]. (Unpublished)

Text 5006211057_5006211058-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Peramalan jumlah realisasi kredit merupakan bagian penting dalam pengambilan keputusan perbankan, khususnya dalam penyusunan strategi kredit yang efisien dan akurat. Untuk itu, diperlukan model peramalan deret waktu yang mampu memproyeksikan nilai kredit di masa mendatang dengan tingkat kesalahan yang minimal. Penelitian ini bertujuan untuk membandingkan kinerja metode Exponential Smoothing dan Autoregressive Integrated Moving Average (ARIMA) dalam meramalkan jumlah realisasi Kredit Pemilikan Rumah (KPR) dan Kredit Agunan Rumah (KAR) pada PT Bank Tabungan Negara Cabang Surabaya. Data yang digunakan merupakan data realisasi kredit bulanan periode Januari 2022 hingga Desember 2023. Penelitian ini menggunakan pendekatan Single Exponential Smoothing (SES), Double Exponential Smoothing (DES), dan model ARIMA dengan evaluasi berdasarkan nilai Mean Absolute Percentage Error (MAPE). Hasil analisis menunjukkan bahwa model ARIMA (0,1,1) memberikan hasil peramalan terbaik untuk kedua jenis kredit, dengan nilai MAPE sebesar 18,83% untuk KPR dan 5,17% untuk KAR. Metode SES menghasilkan MAPE sebesar 32,68771% untuk KAR dan 23,07036% untuk KPR. Dan, metode DES menghasilkan MAPE sebesar 28,72443 % untuk KAR dan sebesar 24,36523% untuk KPR. Dengan demikian, dapat disimpulkan bahwa model ARIMA lebih unggul dalam memproyeksikan jumlah realisasi kredit dibandingkan dengan metode Exponential Smoothing.
=================================================================================================================================
Forecasting the amount of credit realization is an essential component in banking decision-making, particularly in formulating efficient and accurate credit strategies. Therefore, a time series forecasting model is needed to project future credit values with minimal error. This study aims to compare the performance of the Exponential Smoothing method and the Autoregressive Integrated Moving Average (ARIMA) model in forecasting the realization of Home Ownership Loans (KPR) and Home Equity Loans (KAR) at PT Bank Tabungan Negara, Surabaya Branch. The data used in this study are monthly credit realization data from January 2022 to December 2023. The forecasting methods employed include Single Exponential Smoothing (SES), Double Exponential Smoothing (DES), and the ARIMA model, with performance evaluated based on the Mean Absolute Percentage Error (MAPE). The analysis results show that the ARIMA (0,1,1) model provides the best forecast for both credit types, with a MAPE of 18.83% for KPR and 5.17% for KAR. The SES method produced a MAPE of 32.68771% for KAR and 23.07036% for KPR, while the DES method yielded a MAPE of 28.72443% for KAR and 24.36523% for KPR. Thus, it can be concluded that the ARIMA model outperforms the Exponential Smoothing methods in forecasting the credit realization amounts.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	ARIMA, Single Exponential Smoothing, Double Exponential Smoothing, Kredit, Peramalan, Loan, Forecasting
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA280 Box-Jenkins forecasting
Divisions:	Faculty of Mathematics, Computation, and Data Science > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Yola Jovita Silaban
Date Deposited:	24 Jul 2025 04:46
Last Modified:	24 Jul 2025 04:46
URI:	http://repository.its.ac.id/id/eprint/120973

Actions (login required)

View Item