Deteksi Peluang Gagal Bayar Calon Debitur Menggunakan Algoritma Particle Swarm Optimization (PSO) untuk Meningkatkan Kinerja Manajemen Risiko pada Koperasi Simpan Pinjam ABC

Abstract

Savings and Loan Cooperatives (KSP) are financial institutions that have an important role in economic and trade activities, useful for channeling funds in the form of loans to members who need them for business or business. In this paper, we examine the detection of potential debtors' default opportunities using the Particle Swarm Optimization (PSO) algorithm in a logistic regression model. In the analysis method, there are several steps: (1) standardizing the data on the risk factor data of prospective debtors, (2) determining the assumptions of the logistic regression model, (3) estimating the parameters of the logistic regression model using the Particle Swarm Optimization (PSO) algorithm, and (4 ) to test the significance of each variable. The probability of default is determined using the eligibility parameters of the prospective debtor based on past data variables owned by KSP "ABC" in Bandung, Indonesia. The results show that of the eight factors analyzed, there are six factors that have a significant influence on the risk of default, namely the age of the debtor, the number of family dependents, the amount of savings, the amount of collateral, the amount of credit, the credit period with an accuracy of 99.1%. Based on these six factors, a logistic regression model estimator is obtained that can be used to determine the probability of default from prospective debtors. This probability of default is very useful for KSP "ABC" to make a decision on whether or not to give credit, so that the performance of problem loan risk management can be guaranteed.