Financial Distress pada Sektor Industri Manufaktur dengan Metode CART dan BPNN Berdasarkan Model Altman dan Grover

Harsonoseputro, Felicia (2023) Financial Distress pada Sektor Industri Manufaktur dengan Metode CART dan BPNN Berdasarkan Model Altman dan Grover. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06311940000057-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (9MB) | Request a copy

Abstract

Financial distress adalah fenomena kesulitan keuangan yang dianggap darurat keuangan dan merupakan penanda kebangkrutan. Pada tahun 2017, jumlah perusahaan industri di Indonesia mencapai 33.577 dan menurun hingga 30.072 pada tahun 2019. Namun, saat ini Kementerian Investasi dan Badan Koordinasi Penanaman Modal (BKPM) Indonesia tengah mendukung peningkatan investasi di industri manufaktur dalam rangka hilirisasi industri manufaktur oleh pemerintah. Oleh karena itu, dalam meminimalisir kebangkrutan dan mendukung hilirisasi industri manufaktur, klasifikasi financial distress dapat membantu menjadi bahan pertimbangan dengan rasio keuangan dan variabel makroekonomi dapat digunakan sebagai indikasi terjadinya financial distress. Penelitian ini meninjau perusahaan-perusahaan sektor industri manufaktur yang terdaftar di Bursa Efek Indonesia dari kuartal I 2017 hingga kuartal IV 2021 dan berfokus pada perusahaan dengan market capital tertinggi, terendah, dan yang berada di posisi tengah. Kondisi financial distress penelitian ini dilihat dari model Altman dan Grover. Selain itu, penelitian ini membandingkan model dari metode Classification and Regression Tree (CART) dan Backpropagation Neural Network (BPNN). Sebelumnya dilakukan Synthetic Minority Oversampling Technique (SMOTE), apabila terindikasi adanya imbalance data. Variabel prediktor dalam penelitian ini adalah variabel makroekonomi yang berjumlah tiga variabel dan rasio keuangan dengan jumlah 22 rasio yang akan direduksi terlebih dahulu dengan Principal Component Analysis (PCA). Hasil klasifikasi dengan metode CART dan BPNN dibandingkan dengan confusion matrix dalam menemukan metode terbaik. Penelitian ini diharapkan dapat dimanfaatkan oleh perusahaan dan investor sebagai bahan pertimbangan dalam pengambilan keputusan serta pembuatan strategi ataupun langkah tindakan preventif untuk kedepannya. Hasil penelitian mendapatkan metode BPNN sebagai metode terbaik dalam klasifikasi financial distress berdasarkan model Altman dengan model BPNN(8-9-1), yaitu dengan memiliki 8 input neuron, 9 hidden neuron, dan 1 output neuron. Selain itu, metode terbaik dalam klasifikasi financial distress berdasarkan model Grover adalah CART dan BPNN. Kedua metode memiliki hasil analisis confusion matrix yang sama dengan pohon keputusan terbaik memiliki 7 simpul yang 4 diantaranya adalah terminal nodes dengan kedalaman sebanyak 4 dan terjadi pemilahan sebanyak 3 kali. Model BPNN terbaik pada klasifikasi financial distress berdasarkan model Grover adalah BPNN(8-14-1) yang memiliki 8 input neuron, 9 hidden neuron, dan 1 output neuron. Sehingga, ketiga hasil model terbaik dapat digunakan dalam memprediksi kondisi perusahaan mengalami financial distress atau tidak untuk mengantispasi kebangkrutan suatu perusahaan.
================================================================================================================================
Financial distress is a phenomenon of financial difficulty which is considered a financial emergency and is a marker of bankruptcy. In 2017, the number of industrial companies in Indonesia reached 33,577 and decreased to 30,072 in 2019. However, currently the Indonesian Ministry of Investment and the Investment Coordinating Board (BKPM) are supporting increased investment in the manufacturing industry in the context of downstreaming the manufacturing industry by the government. Therefore, in minimizing bankruptcy and supporting the downstreaming of the manufacturing industry, the classification of financial distress can help to be taken into consideration with financial ratios and macroeconomic variables that can be used as indications of financial distress. This research reviews companies in the manufacturing industry sector listed on the Indonesia Stock Exchange from the first quarter of 2017 to the fourth quarter of 2021 and focuses on companies with the highest, lowest, and middle market capital. The condition of financial distress in this study is seen from the Altman and Grover model. In addition, this study compares models from the Classification and Regression Tree (CART) and Backpropagation Neural Network (BPNN) methods. Previously, the Synthetic Minority Oversampling Technique (SMOTE) was performed, if there were indications of data imbalance. The predictor variables in this study are macroeconomic variables, totaling three variables and financial ratios with a total of 22 ratios which will be reduced first by Principal Component Analysis (PCA). Classification results with the CART and BPNN methods are compared with the confusion matrix in finding the best method. This research is expected to be utilized by companies and investors as a consideration in making decisions and developing strategies or preventive action steps for the future. The results showed that the BPNN method is the best method in the classification of financial distress based on the Altman model with the BPNN model (8-9-1), namely by having 8 input neurons, 9 hidden neurons, and 1 output neuron. In addition, the best methods for classifying financial distress based on the Grover model are CART and BPNN. Both methods have the same results of confusion matrix analysis as the best decision tree has 7 nodes, 4 of which are terminal nodes with 4 depths and 3 sorting occurs. The best BPNN model in the financial distress classification based on Grover's model is BPNN (8-14-1) which has 8 input neurons, 9 hidden neurons, and 1 output neuron. Thus, the three best model results can be used in predicting the condition of a company experiencing financial distress or not to anticipate the bankruptcy of a company.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Altman, Backpropagation Neural Network, Classification and Regression Tree, Financial Distress, Grover, Altman, Backpropagation Neural Network, Classification and Regression Tree, Financial Distress, Grover
Subjects:	H Social Sciences > HB Economic Theory H Social Sciences > HG Finance H Social Sciences > HG Finance > HG4012 Mathematical models Q Science > QA Mathematics > QA336 Artificial Intelligence 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:	Felicia Harsonoseputro
Date Deposited:	13 Jul 2023 01:40
Last Modified:	13 Jul 2023 01:40
URI:	http://repository.its.ac.id/id/eprint/98443

Actions (login required)

View Item