Klasifikasi Financial Distress Menggunakan Feedforward Neural Network Berdasarkan Rasio Keuangan Altman dan Ohlson

Pratiwi, Annisa Salsabila (2023) Klasifikasi Financial Distress Menggunakan Feedforward Neural Network Berdasarkan Rasio Keuangan Altman dan Ohlson. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06311940000015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (4MB) | Request a copy

Abstract

Laporan Bank Dunia dengan judul “Is a Global Recession Imminent?’ memprediksi adanya potensi resesi ekonomi global pada tahun 2023 sebagai dampak dari upaya pemulihan ekonomi pasca pandemi Covid-19. Kondisi perekonomian yang terus berubah menuntut perusahaan untuk dapat mengantisipasi kondisi keuangan mendatang demi menghindari terjadinya financial distress, yaitu penurunan kondisi keuangan yang berkelanjutan dan sering terjadi sebelum kebangkrutan. Maka dari itu, penelitian mengenai financial distress perlu dikembangkan sebagai langkah antisipasi dan peringatan dini. Fokus penelitian ini adalah membandingkan rasio keuangan Altman dan Ohlson dalam mengklasifikasikan financial distress pada perusahaan Properti dan Real Estate menggunakan metode Feedforward Neural Network. Penelitian ini menggunakan data laporan keuangan 19 perusahaan Properti dan Real Estate yang terdaftar di Bursa Efek Indonesia mulai dari kuartal I 2016 hingga kuartal III tahun 2022, dengan status awal kondisi keuangan perusahaan adalah berdasarkan Earnings Per Share (EPS). Pada penelitian ini, digunakan juga metode Synthetic Minority Oversampling Technique (SMOTE) sebagai upaya untuk mengatasi imbalance class. Model dan metode terbaik dipilih berdasarkan nilai akurasi dan Area Under Curve (AUC). Rasio keuangan Altman dengan arsitektur model FFNN (5-2-1) melalui proses balancing sebesar 60:40 pada data training menghasilkan akurasi 82,69% dan AUC 0,8204. Rasio keuangan Ohlson dengan proses balancing data 60:40 dan arsitektur model FFNN (9-4-1) menghasilkan akurasi 93,27% dan AUC 0,9045. Dengan demikian, dalam memprediksi financial distress perusahaan Properti dan Real Estate, rasio keuangan Ohlson dengan variabel prediktor Ukuran Perusahaan (SIZE), Total Liabilities to Total Assets (TLTA), Working Capital to Total Assets (WCTA), Current Liabilities to Current Assets (CLCA), Perbandingan Total Liabilities dan Total Assets (OENEG), Net Income to Total Assets (NITA), Cash Flows Operating to Total Liabilities (CFOTL), Nilai Net Income (INTWO), dan Perubahan Net Income (CHIN), memberikan hasil terbaik. Hasil klasifikasi ini dapat digunakan sebagai pertimbangan dalam menggunakan alternatif rasio keuangan untuk mengklasifikasikan financial distress.
================================================================================================================================
The World Bank report titled “Is a Global Recession Imminent?” predicts a potential global economic recession in 2023 as a result of economic recovery efforts after the Covid-19 pandemic. The changing economic conditions require companies to be able to anticipate future financial conditions in order to avoid financial distress, which is a continuous decline in financial condition and often occurs before bankruptcy. Therefore, research on financial distress needs to be developed as an anticipatory step and early warning. The focus of this research is comparing Altman and Ohlson financial ratios in classifying financial distress in Property and Real Estate companies using the Feedforward Neural Network method. This study uses financial statement data of 19 Property and Real Estate companies listed on the Indonesia Stock Exchange from the first quarter of 2016 to the third quarter of 2022, with the initial status of the company’s financial condition based on Earnings Per Share (EPS). In this study, the Synthetic Minority Oversampling Technique (SMOTE) method is also used as an effort to overcome class imbalance. The best model and method are selected based on the accuracy value and Area Under Curve (AUC). Altman’s financial ratios with 60:40 data balancing process and model architecture FFNN (5-2-1) resulted in 82,69% accuracy and 0,8204 AUC. Ohlson’s financial ratios with 60:40 data balancing process and FFNN model architecture (9-4-1) resulted in 93,27% accuracy and 0,9045 AUC. Thus, in predicting financial distress of Property and Real Estate companies, Ohlson’s financial ratio with predictor variables Company Size (SIZE), Total Liabilities to Total Asset (TLTA), Working Capital to Total Assets (WCTA), Current Liabilities to Current Assets (CLCA), Comparison of Total Liabilities and Total Asset (OENEG), Net Income to Total Assets (NITA), Cash Flows Operating to Total Liabilities (CFOTL), Net Income Value (INTWO), and change in Net Income (CHIN) provides the best results. These classification results can be used as a consideration in using alternative financial ratios to classify financial distress.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Altman, Area Under Curve, Feedforward Neural Network, Financial Distress, Ohlson.
Subjects:	H Social Sciences > HG Finance Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q325.78 Back propagation Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Annisa Salsabila Pratiwi
Date Deposited:	27 Jun 2023 01:10
Last Modified:	27 Jun 2023 04:19
URI:	http://repository.its.ac.id/id/eprint/98230

Actions (login required)

View Item