Hidayah, Ifayanti Rohmatul (2024) Pemodelan Financial Distress Pada Perusahaan Sektor Property & Real Estate Menggunakan Pendekatan Explainable Historical Random Forest. Diploma thesis, Institut Teknologi Sepuluh Nopember.
Text
2043201096-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (4MB) | Request a copy |
Abstract
Sektor property & real estate memiliki risiko yang tinggi dan sulit diprediksi membuat sektor ini rentan mengalami financial distress. Financial distress merupakan kondisi ketidakmampuan suatu perusahaan untuk memenuhi kewajiban keuangan seperti membayar utang. Penelitian ini bertujuan untuk mengembangkan sebuah model prediksi financial distress pada 73 perusahaan sektor property & real estate menggunakan metode Historical Random Forest dengan menambahkan pendekatan Variabel Importance dan SMOTE. Metode Historical Random Forest digunakan dalam penelitian ini karena kemampuannya dalam menghasilkan prediksi dengan mempertimbangkan waktu historis, sehingga lebih unggul dibandingkan metode Random Forest. Pendekatan metode SMOTE digunakan untuk menyeimbangkan kelas pada dataset klasifikasi. Metode Variabel Importance berguna untuk menginterpretasikan dan memahami kontribusi variabel prediktor terhadap pembentukan model dalam menjelaskan kemungkinan terjadinya financial distress pada perusahaan. Data yang digunakan penelitian ini menggunakan data sekunder yang diperoleh dari laporan keuangan tahunan pada 73 perusahaan property & real estate yang terdapat di Bursa Efek Indonesia selama tahun 2018-2022. Penelitian ini juga akan membandingkan hasil prediksi metode Historical Random Forest dengan metode Random Forest guna mengetahui apakah benar metode Historical Random Forest dapat memberikan hasil yang lebih baik jika data yang digunakan mempertimbangkan waktu historis. Hasil dari penelitian ini diharapkan dapat memberikan kontribusi dalam pengembangan model prediksi financial distress pada 73 perusahaan sektor property & real estate yang lebih akurat dan dapat diinterpretasikan dengan baik. Model ini akan membantu para pemangku kepentingan, termasuk investor dan manajemen perusahaan dalam mengidentifikasi risiko financial distress secara dini dan mengambil tindakan pencegahan yang tepat untuk menjaga stabilitas dan kelangsungan bisnis perusahaan sektor property & real estate.
=============================================================================================================================
The property & real estate sector has high risks and difficult to predict, making this sector vulnerable to financial distress. Financial distress is a condition of a company's inability to meet financial obligations such as paying debts. This research aims to develop a financial distress prediction model for 73 property & real estate sector companies using the Historical Random Forest method by adding the Variable Importance and SMOTE methods. The Historical Random Forest method is used in this study because of its ability to generate predictions by considering historical time, making it more advanced than the Random Forest method. The SMOTE method approach is used to balance the classes in the classification dataset. The Variable Importance method is useful for interpreting and understanding the contribution of predictor variables to the formation of the model in explaining the possibility of financial distress in the company. The data used in this research uses secondary data obtained from annual financial reports on 73 property & real estate companies listed on the Indonesia Stock Exchange during 2018-2022. This research will also compare the prediction results of the Historical Random Forest method with the Random Forest method to determine whether it is true that the Historical Random Forest method can provide better results if the data used considers historical time. The results of this research are expected to contribute to the development of a financial distress prediction model in 73 property & real estate sector companies that are more accurate and can be interpreted properly. This model will help stakeholders, including investors and company management in identifying the risk of financial distress early and taking appropriate precautions to maintain the stability and business continuity of property & real estate sector companies.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | Financial Distress, Historical Random Forest, Property & Real Estate, Random Forest, SMOTE, Variabel Importance |
Subjects: | R Medicine > R Medicine (General) > R858 Deep Learning T Technology > T Technology (General) > T57.5 Data Processing |
Divisions: | Faculty of Vocational > 49501-Business Statistics |
Depositing User: | Ifayanti Rohmatul Hidayah |
Date Deposited: | 16 Feb 2024 03:09 |
Last Modified: | 16 Feb 2024 03:09 |
URI: | http://repository.its.ac.id/id/eprint/107070 |
Actions (login required)
View Item |