Wikristina, Tirza Malta (2023) Penerapan machine learning untuk memprediksi financial distress pada perusahaan sektor perindustrian di Indonesia = Implementation of machine learning for predicting financial distress in industrials sector companies in Indonesia. Bachelor thesis, Universitas Pelita Harapan.
|
Text (Title)
Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (277kB) | Preview |
|
|
Text (Abstract)
Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (730kB) | Preview |
|
|
Text (ToC)
ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (821kB) | Preview |
|
|
Text (Chapter1)
Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (719kB) | Preview |
|
Text (Chapter2)
Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB) |
||
Text (Chapter3)
Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (807kB) |
||
Text (Chapter4)
Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB) |
||
Text (Chapter5)
Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (710kB) |
||
|
Text (Bibliography)
Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (706kB) | Preview |
|
Text (Appendices)
Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB) |
Abstract
Kebangkrutan merupakan persoalan serius yang memakan begitu banyak biaya. Tidak hanya merugikan pemilik perusahaan, kebangkrutan juga merugikan para karyawan, investor, kreditur, bahkan kestabilan perekonomian suatu negara. Maka dari itu untuk mendapatkan early warning yang mendeteksi potensi kebangkrutan sejak awal, dilakukan prediksi terhadap financial distress. Secara umum, financial distress didefinisikan sebagai kondisi di saat keuangan perusahaan sedang tidak baik, sehingga perusahaan kesulitan atau bahkan tidak mampu untuk memenuhi kewajibannya. Dengan menggunakan data rasio keuangan, penelitian ini akan menggunakan machine learning: Support Vector Machine (SVM), Logistic Regression (LR), dan kombinasi antara keduanya. Total terdapat enam model yang dibentuk, yaitu: model LR, LR-LR, LR-BW, SVM, SVM-LR, dan SVM-BW. Pada penelitian ini keenam model digunakan untuk memprediksi empat kelas financial distress, yaitu: tidak financial distress, Profit Reduction (PR), Mild Liquidity (ML), dan Severe Liquidity (SL). Performa dari keenam model akan dibandingkan berdasarkan nilai weighted accuracy, weighted F1 score, dan weighted Area Under the Precision-Recall Curve (AUC-PR). Hasil yang didapatkan dari penelitian ini menunjukkan bahwa model Support Vector Machine yang menerapkan seleksi fitur eliminasi mundur atau model SVM-BW memiliki performa yang paling baik dalam memprediksi financial distress dengan weighted accuracy 74,5%; weighted F1 score 70,6%; dan weighted AUC-PR 79,6%. / Bankruptcy is a serious problem with significant costs. It harms not only the owners of the company, but also its employees, investors, creditors, and even the stability of a country's economy. Therefore, financial distress is predicted in order to provide an early warning of potential bankruptcy. Generally, financial distress is defined as a condition in which a company's finances are not good, making it difficult or even impossible for it to meet its obligations. Using financial ratio data, this study will use machine learning: Support Vector Machine (SVM), Logistic Regression (LR), and a combination of both. A total of six models will be formed, namely LR, LR-LR, LR-BW, SVM, SVM-LR, and SVM-BW. In this study, all six models are used to predict four classes of financial distress, namely, no financial distress, Profit Reduction (PR), Mild Liquidity (ML), and Severe Liquidity (SL). The performance of the six models is compared based on weighted accuracy, weighted F1 score, and weighted Area Under the Precision-Recall Curve (AUC-PR). The results obtained from this study indicate that the Support Vector Machine model that applies backward elimination feature selection or the SVM-BW model has the best performance in predicting financial distress with weighted accuracy 74.5 %; weighted F1 score 70.6%; and weighted AUC-PR 79.6%.
Item Type: | Thesis (Bachelor) | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Creators: |
|
||||||||||||
Contributors: |
|
||||||||||||
Uncontrolled Keywords: | support vector machine; logistic regression; klasifikasi; multiclass; financial distress | ||||||||||||
Subjects: | Q Science > QA Mathematics | ||||||||||||
Divisions: | University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics |
||||||||||||
Depositing User: | Tirza Malta Wikristina | ||||||||||||
Date Deposited: | 27 Jul 2023 07:37 | ||||||||||||
Last Modified: | 27 Jul 2023 07:37 | ||||||||||||
URI: | http://repository.uph.edu/id/eprint/57051 |
Actions (login required)
View Item |