Peramalan Jumlah Kasus Kredit Bermasalah di Koperasi XYZ dengan Menggunakan Model Poisson Generalized Autoregressive Moving Average (GARMA)

Simanjuntak, David Mikhael (2023) Peramalan Jumlah Kasus Kredit Bermasalah di Koperasi XYZ dengan Menggunakan Model Poisson Generalized Autoregressive Moving Average (GARMA). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111940000102-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Kredit bermasalah merupakan suatu kondisi dimana terjadinya cidera janji dalam repayment kredit yang mengakibatkan adanya tunggakan atau potensi kerugian sehingga memungkinkan timbulnya risiko pada usaha kreditur di kemudian hari. Berdasarkan publikasi yang dilakukan oleh Otoritas Jasa Keuangan (OJK), jumlah kredit bermasalah di Indonesia mengalami peningkatan yang signifikan dan mengalami fluktuasi dalam beberapa tahun terakhir. Hal ini sejalan dengan peningkatan jumlah kasus kredit bermasalah koperasi di Kota Tebing Tinggi. Berdasarkan hal ini perlu dilakukan peramalan pada data jumlah kasus kredit bermasalah koperasi di Kota Tebing Tinggi. Data jumlah kasus kredit bermasalah merupakan data count. Salah satu model peramalan yang umum digunakan yaitu Autoregressive Integrated Moving Average (ARIMA), namun metode ini tidak selalu tepat digunakan untuk data count. Generalized Linear Model (GLM) merupakan solusi yang digunakan untuk menganalisis data count. Penelitian terus dilakukan dan menghasilkan pengembangan model peramalan yaitu Generalized Autoregressive Moving Average (GARMA) untuk data yang mengikuti distribusi non–Gaussian seperti distribusi Poisson. Dalam mengestimasi parameter model Poisson GARMA digunakan metode Maximum Likelihood Estimation (MLE) dengan pendekatan optimasi Iteratively Reweighted Least Square (IRLS). Hasil yang didapat pada penelitian ini adalah model Poisson GARMA (1,2) merupakan model terbaik untuk peramalan data tersebut dengan pemilihan model terbaik berdasarkan nilai RMSE dan MAPE terkecil. Nilai RMSE Poisson GARMA (1,2) sebesar 26,17369640 dan MAPE sebesar 16%. Poisson GARMA (1,2) diimplementasikan pada data jumlah kasus kredit bermasalah koperasi di Kota Tebing Tinggi dan didapatkan hasil peramalan pada tahun 2023 cenderung mengalami penurunan jumlah kasus dari bulan Januari – April 2023 dan peningkatan jumlah kasus dari bulan April – Desember 2023.
=====================================================================================================================================
Non-performing loans are a condition where there is a breach of promise in credit repayment which results in arrears or potential losses, allowing risks to the creditor's business in the future. Based on publications by the Otoritas Jasa Keuangan (OJK), the number of non-performing loans in Indonesia has increased significantly and has fluctuated in recent years. This is in line with the increase in the number of union non-performing loan cases in Tebing Tinggi City. Based on this, it is necessary to forecast the data on the number of cases of union non-performing loans in Tebing Tinggi City. Data on the number of non-performing loan cases is count data. One of the commonly used forecasting models is Autoregressive Integrated Moving Average (ARIMA), but this method is not always appropriate for count data. Generalized Linear Model (GLM) is a solution used to analyze count data. Research continues and results in the development of a forecasting model, namely Generalized Autoregressive Moving Average (GARMA) for data that follows a non-Gaussian distribution such as the Poisson distribution. To estimate the parameters of the Poisson GARMA model, the Maximum Likelihood Estimation (MLE) method is used with the Iteratively Reweighted Least Square (IRLS) optimization approach. The results obtained in this research are that the Poisson GARMA (1,2) model is the best model for forecasting these data with the selection of the best model based on the smallest RMSE and MAPE values. The RMSE value of Poisson GARMA (1,2) is 26,17369640 and MAPE is 16%. Poisson GARMA (1,2) of 26.17369640 and MAPE of 16%. Poisson GARMA (1,2) is implemented on data on the number of cases of cooperative non-performing loans in Tebing Tinggi City and obtained forecasting results in 2023 tend to experience a decrease in the number of cases from January - April 2023 and an increase in the number of cases from April - December 2023.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kredit Bermasalah, Data Count, Poisson GARMA, IRLS, Non-Performing Loans, Count Data, Poisson GARMA, IRLS
Subjects:	H Social Sciences > HG Finance H Social Sciences > HJ Public Finance
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	David Mikhael Simanjuntak
Date Deposited:	08 Aug 2023 06:11
Last Modified:	08 Aug 2023 06:11
URI:	http://repository.its.ac.id/id/eprint/103125

Actions (login required)

View Item