Perbandingan Model ARIMA dan Support Vector Regression untuk Peramalan Pinjaman Bank (BRI, Mandiri, BNI, BCA)

Pratisto, Yohanes Kristianto (2023) Perbandingan Model ARIMA dan Support Vector Regression untuk Peramalan Pinjaman Bank (BRI, Mandiri, BNI, BCA). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211940000133-Undegraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Pinjaman erat kaitannya dengan berjalannya perekonomian dalam suatu negara. Sehubungan dengan itu, bank adalah salah satu sumber yang memberikan pinjaman terhadap masyarakat, sehingga diperlukan peramalan untuk melihat besarnya pinjaman yang akan dikeluarkan oleh bank di masa depan. Selain untuk melihat besarnya pinjaman yang akan dikeluarkan oleh bank, peramalan tersebut juga dapat menjadi acuan untuk menciptakan produk-produk perbankan yang berkaitan dengan pinjaman. Peramalan data pinjaman bank dilakukan terhadap empat bank terbesar di Indonesia antara lain Bank Rakyat Indonesia (BRI), Bank Mandiri, Bank Negara Indonesia (BNI), dan Bank Central Asia (BCA). Diperlukan metode peramalan yang akurat untuk melihat proyeksi pinjaman bank di masa depan sehingga pada penelitian ini dilakukan peramalan menggunakan perbandingan dua metode, antara lain ARIMA dan Support Vector Regression (SVR). ARIMA merupakan salah satu metode peramalan berbasis distribusi normal. Sedangkan SVR merupakan salah satu machine learning yang bebas dari asumsi distribusi. Data yang digunakan adalah data pinjaman per bulan dari BRI, Bank Mandiri, BNI, dan BCA. Model terbaik yang didapatkan untuk peramalan pinjaman di BRI, Mandiri, BNI dan BCA adalah ARIMA dengan MAPE outsample masing masing berturut-turut sebesar 0,82%;1,21%;1,09%;0,86%.
===============================================================================================================================
Loans are closely related to the running of the economy within a country. In this regard, the bank is one of the sources that provides loans to the public, so forecasting is needed to see the amount of loans that will be issued by the bank in the future. In addition to seeing the amount of loans to be issued by banks, the forecasting can also be a reference for creating banking products related to loans. Forecasting the volume of bank loan is carried out on the four largest banks in Indonesia, including Bank Rakyat Indonesia (BRI), Bank Mandiri, Bank Negara Indonesia (BNI), and Bank Central Asia (BCA). An accurate forecasting tool is needed to see the projection of bank loans in the future, so that in this study forecasting was carried out using the comparison of three methods, including ARIMA and Support Vector Regression (SVR). ARIMA is one of the normal distribution-based forecasting methods. Meanwhile, SVR is one of the machine learning that is free from distribution assumptions. The data used are monthly bank loan for BRI, Bank Mandiri, BNI, and BCA. The best model obtained for loan forecasting at BRI, Mandiri, BNI, and BCA are ARIMA with MAPE out sample of 0,82%;1,21%;1,09%;0,86%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	ARIMA, Bank, Loan, Support Vector Regression, Bank, Pinjaman
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HB Economic Theory > Economic forecasting--Mathematical models. H Social Sciences > HG Finance Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Yohanes Kristianto Pratisto
Date Deposited:	06 Sep 2023 02:53
Last Modified:	06 Sep 2023 02:53
URI:	http://repository.its.ac.id/id/eprint/104342

Actions (login required)

View Item