Penerapan Support Vector Machine Dengan SMOTE Pada Analisis Prediksi Financial Distress di Sektor Transportasi & Logistik

Salma, Rahmi Waladiyatusalma (2024) Penerapan Support Vector Machine Dengan SMOTE Pada Analisis Prediksi Financial Distress di Sektor Transportasi & Logistik. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2043201052_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (15MB) | Request a copy

Abstract

Sektor transportasi dan logistik memiliki peran penting sebagai penghubung kegiatan ekonomi dan perdagangan antar wilayah. Oleh karena itu, perusahaan pada sektor ini diharapkan dapat menjaga kondisi keuangan perusahaan agar terhindar dari permasalahan keuangan seperti financial distress. Financial distress adalah tahap penurunan kondisi keuangan yang terjadi sebelum kebangkrutan. Tujuan dilakukan penelitian ini yaitu untuk mengetahui keakuratan hasil prediksi klasifikasi financial distress pada perusahaan sektor transportasi dan logistik menggunakan metode Support Vector Machine dengan Synthetic Minority Oversampling Technique (SMOTE). Data yang digunakan dalam penelitian ini merupakan rasio keuangan perusahaan sektor transportasi dan logistik yang terdaftar di Bursa Efek Indonesia pada tahun 2018 - 2021 dengan jumlah Perusahaan yang digunakan sebanyak 21 perusahaan pada setiap tahunnya. Variabel prediktor yang digunakan merupakan 9 nilai rasio keuangan. Untuk klasifikasi kategori kondisi awal perusahaan menggunakan rasio keuangan Return of Total Asset, Debt to Asset Ratio dan Current Ratio. Sedangkan untuk klasifikasi dengan metode Support Vector Machine menggunakan rasio keuangan Quick Ratio, Working Capital to Total Asset, Total Asset Turnover, Inventory Turnover, Net Profit Margin dan Earning Per Share. Hasil penelitian menunjukkan berdasarkan karakteristik data rasio keuangan, kinerja perusahaan sektor transportasi dan logistik di Indonesia berfluktuatif, dengan beberapa perusahaan mengalami kerugian yaitu PT. Air Asia Indonesia Tbk dan beberapa perusahaan memiliki kinerja yang sangat baik seperti PT. Satria Antaran Prima Tbk, PT. Pelayaran Nelly Dwi Putri Tbk dan PT. Trimuda Nuansa Citra Tbk. Selain itu, Proporsi jumlah perusahaan yang mengalami financial distress dan non-financial distress sebelum dilakukan analisis SMOTE pada data training sebesar 23% dan 77%, setelah dilakukan analisis SMOTE proporsi nya menjadi hampir seimbang yaitu financial distress sebesar 54% dan non-financial distress sebesar 46%hasil klasifikasi model SVM kernel Radial Basis Function yang melalui proses balancing menggunakan SMOTE dengan parameter C=5 dan γ=0.001 menghasilkan nilai akurasi 80,95% yang artinya model mampu memprediksi keadaan perusahaan secara keseluruhan sebesar 80,95%, nilai sensitivitas sebesar 71,43% yang artinya model mampu memprediksi perusahaan yang mengalami financial distress sebesar 71,43%, nilai spesifikasi sebesar 85,71% yang artinya model mampu memprediksi perusahaan non-financial distress sebesar 85,71%. Serta, diperoleh nilai AUC sebesar 78,57% yang menunjukkan bahwa performa model dalam melakukan klasifikasi termasuk kedalam kategori cukup baik.
=================================================================================================================================
The transportation and logistics sector has an important role as a link between economic and trade activities between regions. Therefore, companies in this sector are expected to be able to maintain the company's financial condition to avoid financial problems such as financial distress. Financial distress is the stage of decline in financial conditions that occurs before bankruptcy. The aim of this research is to determine the accuracy of the prediction results for financial distress classification in transportation and logistics sector companies using the Support Vector Machine method with Synthetic Minority Oversampling Technique (SMOTE). The data used in this research are the financial ratios of transportation and logistics sector companies listed on the Indonesia Stock Exchange in 2018 - 2021 with a total of 21 companies used each year. The predictor variables used are 9 financial ratio values. To classify the company's initial condition categories, the financial ratios Return of Total Assets, Debt to Asset Ratio and Current Ratio are used. Meanwhile, classification using the Support Vector Machine method uses the financial ratios Quick Ratio, Working Capital to Total Assets, Total Asset Turnover, Inventory Turnover, Net Profit Margin and Earning Per Share. The research results show that based on the characteristics of financial ratio data, the performance of transportation and logistics sector companies in Indonesia fluctuates, with several companies experiencing losses, namely PT. Air Asia Indonesia Tbk and several companies have excellent performance such as PT. Satria Antaran Prima Tbk, PT. Pelayaran Nelly Dwi Putri Tbk and PT. Trimuda Nuansa Citra Tbk. Apart from that, the proportion of companies experiencing financial distress and non-financial distress before carrying out the SMOTE analysis on training data was 23% and 77%, after carrying out the SMOTE analysis the proportion became almost equal, namely financial distress amounting to 54% and non-financial distress amounting to 46% of the classification results of the SVM kernel Radial Basis Function model which went through a balancing process using SMOTE with parameters C=5 and γ=0.001 produced an accuracy value of 80.95%, which means the model was able to predict the overall state of the company by 80.95%, a sensitivity value of 80.95%. 71.43%, which means the model is able to predict companies
experiencing financial distress by 71.43%, the specification value is 85.71%, which means the model is able to predict non-financial distress companies by 85.71%. Also, an AUC value of 78.57% was obtained, which shows that the model's performance in classifying is in the quite good category.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Financial Distress, SMOTE, Support Vector Machine, Transportation and logistic, Financial Distress
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Rahmi Waladiyatusalma
Date Deposited:	05 Feb 2024 03:19
Last Modified:	05 Feb 2024 03:19
URI:	http://repository.its.ac.id/id/eprint/106031

Actions (login required)

View Item