Prediksi Financial Distress Pada Perusahaan Sektor Transportasi di Indonesia Dengan SMOTE-Regresi Logistik dan SMOTE-Regresi Logistik Eksak

Wirastri, Pramesti Hayu (2023) Prediksi Financial Distress Pada Perusahaan Sektor Transportasi di Indonesia Dengan SMOTE-Regresi Logistik dan SMOTE-Regresi Logistik Eksak. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211940000011-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (3MB) | Request a copy

Abstract

Selama masa pandemi Covid-19 di Indonesia banyak perusahaan transportasi yang mengalami tekanan. Seperti diketahui, diberlakukannya kebijakan pemerintah seperti PSBB menyebabkan mobilitas masyarakat menurun secara signifikan, sehingga mempengaruhi pendapatan perusahaan terutama sektor transportasi. Tujuan dari penelitian ini adalah untuk mengetahui apakah rasio keuangan dapat digunakan untuk memprediksi kondisi pada perusahaan sektor transportasi di Indonesia. Pengukuran financial distress menggunakan metode Altman Z-score. Perusahaan dengan nilai lebih dari 1,1 dikategorikan sebagai perusahaan non-financial distress sementara perusahaan dengan nilai kurang dari 1,1 dikategorikan sebagai perusahaan yang mengalami financial distress. Dilakukan pemodelan regresi logistik biner dan regresi logistik eksak untuk mengetahui variabel apa saja yang memiliki pengaruh terhadap financial distress. Variabel independen yang digunakan adalah rasio keuangan yang terdiri dari 20 variabel, sedangkan variabel dependen yang digunakan yaitu “0” untuk kategori non-financial distress, “1” untuk kategori financial distress. Data yang digunakan adalah data sekunder yang diperoleh dari publikasi yang dikeluarkan Bursa Efek Indonesia melalui situs www.idx.co.id. Periode data penelitian ini mencakup data tahun 2021 yang dirasa cukup untuk mewakili dalam memprediksi financial distress karena pada periode tersebut tekanan bagi perusahaan-perusahaan cukup besar pasca terjadinya Covid-19 pada tahun 2020. Pada penelitian ini dilakukan penanganan imbalanced data dengan metode Synthetic Minority Oversampling Technique (SMOTE) karena komposisi data yang tidak seimbang. Didapatkan hasil bahwa variabel yang memiliki pengaruh signifikan adalah Earning/Debt, MVE/Total Liabilities, dan EBIT/Sales. Secara umum, metode SMOTE-regresi logistik memiliki AUC dan sensitivitas yang lebih baik dalam memprediksi financial distress daripada regresi logistik, regresi logistik eksak dan SMOTE-regresi logistik eksak khususnya pada ukuran sampel yang semakin besar.
===============================================================================================================================
During the Covid-19 pandemic in Indonesia, many transportation companies were under pressure. As is known, the enactment of government policies such as the PSBB has caused people's mobility to decrease significantly, thereby affecting company revenues, especially the transportation sector. This research aims to determine whether financial ratios can be used to predict conditions in transportation sector companies in Indonesia. Measurement of financial distress using the Altman Z-score method. Companies with a score of more than 1,1 are categorized as non-financial distress companies while companies with a value of less than 1,1 are categorized as companies experiencing financial distress. Binary logistic regression and exact logistic regression were carried out to find out which variables have an influence on financial distress. The independent variable used is financial ratios consisting of 20 variables, while the dependent variable used is "0" for the category of non-financial distress, "1" for the category of financial distress. The data used is secondary data obtained from publications issued by the Indonesia Stock Exchange through the website www.idx.co.id. The data period for this research includes data for 2021 which is considered sufficient to represent in predicting financial distress because during this period the pressure for companies was quite large after the occurrence of Covid-19 in 2020. In this study, In this study, the handling of imbalanced data was carried out with the Synthetic Minority Oversampling Technique (SMOTE) method due to unbalanced data composition. The results show that the variable that has a significant effect are Earning/Debt, MVE/Total Liabilities, and EBIT/Sales. In general, the SMOTE-logistic regression method has a better AUC and sensitivity in predicting financial distress than logistic regression, exact logistic regression and SMOTE-exact logistic regression especially on increasingly large sample sizes.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Altman Z-Score, Financial Distress, Rasio Keuangan, SMOTE-Regresi Logistik, SMOTE-Regresi Logistik Eksak, Altman Z-score, Financial Distress, Financial Ratio, SMOTE-Binary Logistic Regression, SMOTE-Exact Logistic Regression
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Pramesti Hayu Wirastri
Date Deposited:	20 Feb 2023 05:09
Last Modified:	20 Feb 2023 05:09
URI:	http://repository.its.ac.id/id/eprint/97635

Actions (login required)

View Item