Analisis Perbandingan Klasifikasi Financial Distress Perusahaan Menggunakan Artificial Neural Network, Support Vector Machine, dan Hybrid ANN-SVM pada Perusahaan Retail Tahun 2020-2023

MARISTY WIDYA PANGESTIKA, Dr. Adhitya Ronnie Effendie, S.Si., M.Si., M.Sc.

2024 | Skripsi | STATISTIKA

Financial distress merupakan kondisi penurunan keuangan perusahaan sebelum terjadinya kebangkrutan. Klasifikasi financial distress penting untuk dilakukan karena bermanfaat untuk perusahaan, investor, dan kreditur. Variabel yang digunakan merupakan rasio keuangan pada Altman Z-Score yaitu Working Capital/Total Assets, Retained Earning/Total Assets, Earning Before Interest and Tax/Total Assets, dan Book Value of Equity/Total Liabilities. Pada penelitian ini digunakan data rasio keuangan perusahaan retail yang terdaftar di Bursa Efek Indonesia tahun 2020-2023. Penelitian ini menggunakan metode Artificial Neural Network (ANN), Support Vector Machine (SVM), dan hybrid ANN-SVM. Berdasarkan hasil perbandingan evaluasi performa keseluruhan model untuk melakukan analisis klasifikasi perusahaan financial distress subsektor retail diperoleh bahwa metode hybrid ANN-SVM memberikan nilai tertinggi untuk akurasi, presisi, sensitivitas, dan F1 Score.

Financial distress is a phase of a decline in financial conditions occurred before the occurrence of bankruptcy. Classification of financial distress are important for the company, investors, creditors, etc. The variables used in this study are financial ratios on the Altman Z-Score, namely Working Capital/Total Assets, Retained Earning/Total Assets, Earning Before Interest and Tax/Total Assets, and Book Value of Equity/Total Liabilities. The population in this study are the financial ratios of retail companies listed on the Indonesia Stock Exchange in 2020-2023. The method used in this research is Support Vector Machine (SVM), Artificial Neural Network (ANN),and hybrid ANN-SVM. Based on the results of the comparing performance evaluation of the model such as accuracy, precision, sensitivity, specificity, and F1 Score, the hybrid ANN-SVM model obtained the best performance.

Kata Kunci : financial distress, ANN, SVM, hybrid ANN-SVM