Prediksi Financial Distress Perusahaan Berdasarkan Rasio-Rasio Keuangan Dengan Menggunakan Support Vector Machine (SVM)-SMOTE

Wahyudi, Rizzaqi Khoirul (2024) Prediksi Financial Distress Perusahaan Berdasarkan Rasio-Rasio Keuangan Dengan Menggunakan Support Vector Machine (SVM)-SMOTE. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 09211950095006-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (3MB) | Request a copy

Abstract

Financial Distress merupakan kondisi kesulitan keuangan yang dialami oleh perusahaan dan dapat berujung pada kebangkrutan. Kondisi ini disebabkan oleh beberapa faktor, baik faktor internal maupun eksternal dari perusahaan terutama adanya krisis ekonomi. Beberapa krisis ekonomi pernah dirasakan di Indonesia, antara lain pada tahun 1997 (Asian Financial Crisis), tahun 2008 (Global Financial Crisis), dan tahun 2020 (Covid-19 Outbreak). Manajemen perusahaan perlu memahami kondisi perusahaan yang sudah menunjukkan kondisi financial distress agar dapat segera mengambil keputusan strategis yang dapat mencegah kebangkrutan. Tujuan dari penelitian ini untuk mengembangkan model prediksi dalam mengidentifikasi potensi financial distress sehingga mampu menjadi early warning system yang dapat mendorong pengambilan keputusan strategis perusahaan dan penentuan keputusan investasi agar terhindar dari risiko kerugian. Penelitian ini menginvestigasi prediksi financial distress berdasarkan rasio – rasio keuangan dengan menggunakan metode Support Vector Machine (SVM) – SMOTE. Sampel yang digunakan dalam penelitian ini sebanyak 95 perusahaan yang terdaftar di Bursa Efek Indonesia (BEI) tahun 2015-2021 dibidang manufaktur dengan jenis rasio-rasio keuangan yang digunakan adalah rasio likuiditas, rasio profitabilitas, rasio leverage dan rasio aktivitas. Perusahaan yang mengalami financial distress diukur berdasarkan Interest Coverage Ratio (ICR) yang menunjukkan kemampuan perusahaan untuk menghasilkan cukup laba untuk membayar biaya bunganya sendiri. SMOTE bertujuan untuk menyeimbangkan data pada variabel distress yang sebelumnya memiliki perbandingan yang sangat jauh 98:567 dan setelah dilakukan SMOTE data memiliki keseimbangan yang baik. Hasil penelitian menunjukkan metode Support Vector Machine (SVM) – SMOTE dalam memprediksi financial distress perusahaan dapat menghasilkan tingkat akurasi mencapai 92.48% berdasarkan pengujian kebaikan model sehingga dapat dengan baik memprediksi kondisi suatu perusahaan tergolong distress atau tidak distress untuk selanjutnya digunakan sebagai pertimbangan pengambilan keputusan strategis.
=================================================================================================================================
Financial Distress is a condition of financial difficulty experienced by a company and can lead to bankruptcy. This condition is caused by several factors, both internal and external to the company, especially the economic crisis. Several economic crises have been experienced in Indonesia, including in 1997 (Asian Financial Crisis), 2008 (Global Financial Crisis), and 2020 (Covid-19 Outbreak). Company management needs to understand the condition of companies that are showing financial distress so they can immediately make strategic decisions that can prevent bankruptcy. The aim of this research is to develop a predictive model to identify potential financial distress so that it can become an early warning system that can encourage company strategic decision making and investment decisions to avoid the risk of loss. This research investigates the prediction of financial distress based on financial ratios using the Support Vector Machine (SVM) - SMOTE method. The sample used in this research was 95 companies listed on the Indonesia Stock Exchange (BEI) in 2015-2021 in the manufacturing sector with the types of financial ratios used being liquidity ratios, profitability ratios, leverage ratios and activity ratios. Companies experiencing financial distress are measured based on the Interest Coverage Ratio (ICR), which shows the company's ability to generate enough profit to pay its own interest costs. SMOTE aims to balance the data on the distress variable which previously had a very large ratio of 98:567 and after SMOTE was carried out the data had a good balance. The results of the research show that the Support Vector Machine (SVM) - SMOTE method in predicting company financial distress can produce an accuracy level of 92.48% based on testing the goodness of the model so that it can properly predict whether a company is classified as distressed or not distressed and then used as a consideration for strategic decision making.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Financial Distress, Support Vector Machine, SMOTE
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT)
Depositing User:	RIZZAQI KHOIRUL WAHYUDI
Date Deposited:	01 Mar 2024 04:12
Last Modified:	01 Mar 2024 04:12
URI:	http://repository.its.ac.id/id/eprint/107746

Actions (login required)

View Item