Aileen, Aurelia Wong (2024) Analisis perbandingan prediksi financial distress menggunakan lasso logistic regression, naive bayes, dan random forest = Comparative analysis of financial distress prediction using lasso logistic regression, naïve bayes, and random forest. Bachelor thesis, Universitas Pelita Harapan.
Text (Title)
Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (158kB) |
|
Text (Abstract)
Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (301kB) |
|
Text (ToC)
ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (568kB) |
|
Text (Chapter1)
Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (298kB) |
|
Text (Chapter2)
Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (639kB) |
|
Text (Chapter3)
Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (597kB) |
|
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 (240kB) |
|
Text (Bibliography)
Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (257kB) |
|
Text (Appendices)
Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB) |
Abstract
Financial distress merujuk pada kondisi di mana terjadi penurunan dalam keadaan keuangan suatu perusahaan, yang muncul sebelum mencapai tahap kebangkrutan atau likuidasi. Oleh karena itu penting sekali memprediksi financial distress dalam perusahaan agar dapat mengatasi secepat mungkin masalah financial distress. Studi ini membentuk dan membandingkan hasil klasifikasi financial distress menggunakan tiga metode berbeda, yaitu Random Forest, Naïve Bayes, dan LASSO logistic regression dengan menggunakan data rasio keuangan perusahaan. Ketiga metode ini dibentuk menjadi enam model dan akan diuji dalam penelitian ini, termasuk model Random Forest, Naïve Bayes, dan LASSO logistic regression yang menggunakan semua variabel rasio keuangan, serta model yang menggunakan seleksi variabel dari LASSO regression. Keenam model tersebut dievaluasi terhadap dua kelas (biner) berisikan kelas 0 (financial distress serta kelas 1 (sehat) dan tiga kelas (multiclass) berisikan kelas 0 (financial distress), kelas 1 (zona abu - abu), serta kelas 2 (sehat). Hasil analisis dari penelitian ini menunjukkan bahwa penggunaan variabel seleksi dengan LASSO regression memiliki dampak negatif terhadap kinerja model Random Forest dan Naïve Bayes pada klasifikasi multiclass. Namun, perbedaan kinerja antara model menggunakan semua variabel dan model dengan seleksi variabel tidak signifikan. Model Random Forest dengan semua variabel menunjukkan kinerja terbaik dengan akurasi sebesar 0,9726, weighted F1 Score sebesar 0,9722, AUC-PR sebesar 0,9185. Faktor-faktor yang mempengaruhi adalah EBIT Margin, Net Income Margin, Return On Assets, Return On Equity, Total Debt to Equity, Current Ratio, Days Payable Outstanding, Fixed Assets Turnover, dan Total Asset Turnover. / Financial distress refers to a condition where there is a decline in the financial condition of a company, which appears before it reaches the bankruptcy or liquidation stage. Therefore, it is very important to predict financial distress in a company so that it can resolve financial distress problems as quickly as possible. This study forms and compares the results of financial distress classification using three different methods, namely Random Forest, Naïve Bayes, and LASSO logistic regression using company financial ratio data. These three methods were formed into six models and will be tested in this research, including the random forest, Naïve Bayes, and LASSO logistic regression models. using all financial ratio variables, as well as a model that uses variable selection from LASSO regression. The six models were evaluated against two classes (binary) containing class 0 (financial distress and class 1 (healthy) and three classes (multiclass) containing class 0 (financial distress), class 1 (gray zone), and class 2 (healthy). The analysis results from this research show that using selection variables with LASSO regression has a negative impact on the performance of the Random Forest and Naïve Bayes models in multiclass classification. However, the difference in performance between the model using all variables and the model with variable selection is not significant. Random Forest model with all variables showed the best performance with an accuracy of 0,9726, Weighted F1 Score of 0,9722, and an AUC-PR of 0,9185. The influencing factors are EBIT Margin, Net Income Margin, Return On Assets, Return On Equity, Total Debt to Equity, Current Ratio, Days Payable Outstanding, Fixed Assets Turnover, and Total Asset Turnover.
Item Type: | Thesis (Bachelor) | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Creators: |
|
||||||||||||
Contributors: |
|
||||||||||||
Uncontrolled Keywords: | financial distress; lasso logistic regression; random forest; naïve bayes; rasio keuangan; financial distress; lasso regression; random forest; naïve bayes financial ratio. | ||||||||||||
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: | Aurelia Wong Aileen | ||||||||||||
Date Deposited: | 16 Jul 2024 01:12 | ||||||||||||
Last Modified: | 16 Jul 2024 01:12 | ||||||||||||
URI: | http://repository.uph.edu/id/eprint/63993 |
Actions (login required)
View Item |