Prediksi Financial Distress pada Perusahaan Sektor Pertambangan di Indonesia dengan Pendekatan Deep Learning (DL)

Zulfikar, Marcia Nayfa Fahira (2024) Prediksi Financial Distress pada Perusahaan Sektor Pertambangan di Indonesia dengan Pendekatan Deep Learning (DL). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201133-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (2MB) | Request a copy

Abstract

Selama beberapa dekade, isu mengenai financial distress telah menjadi fokus kajian mendalam di bidang keuangan perusahaan. Dampaknya yang signifikan terhadap kelangsungan hidup bisnis, serta keputusan investor dan kreditur eksternal, menjadikan pemahaman dan prediksi financial distress sangat penting. Pentingnya memprediksi financial distress perusahaan dengan akurat, terutama di sektor pertambangan, menjadi krusial mengingat kontribusinya yang signifikan terhadap pertumbuhan ekonomi Indonesia. Sebagaimana sektor pertambangan dan penggalian memberikan kontribusi sebesar 12,22% terhadap pertumbuhan ekonomi nasional di tahun 2022. Memastikan prediksi yang tepat terkait financial distress menjadi kunci dalam memitigasi risiko dan menjaga stabilitas sektor ini. Dengan mengintegrasikan metode deep learning, yaitu Artificial Neural Network (ANN), khususnya One-Dimensional Convolutional Neural Network (1D-CNN). Dalam penelitian ini, digunakan variable prediktor berupa 11 rasio keuangan dan 1 variabel respon berupa klasifikasi biner melalui pendekatan nilai indeks Interest Coverage Ratio (ICR). Data dibagi menjadi 80% untuk training dan 20% untuk testing dengan Stratified 10-fold Cross Validation. Dengan tujuan untuk mendapatkan model dengan akurasi terbaik sebagai early warning perusahaan, model terbaik yang diperoleh dalam penelitian ini adalah model dengan kernel size 3, learning rate α 0,0001, dan regularization L2 pada 3 layer terpilih dengan nilai akurasi sebesar 92,31%.
======================================================================================================================================================================================================================
Over the decades, the issue of financial distress has been a deep focus of study in corporate finance. Its significant impact on business continuity, as well as the decisions of investors and external creditors, makes understanding and predicting financial distress crucial. Accurately predicting financial distress of companies, especially in the mining sector, is vital given its substantial contribution to Indonesia's economic growth. The mining and quarrying sector contributed 12.22% to national economic growth in 2022. Ensuring precise predictions regarding financial distress is key to mitigating risks and maintaining the stability of this sector. By integrating deep learning methods, particularly Artificial Neural Networks (ANN), specifically One-Dimensional Convolutional Neural Networks (1D-CNN), this study uses 11 financial ratios as predictor variables and one binary classification response variable through the Interest Coverage Ratio (ICR) index value approach. The data is divided into 80% for training and 20% for testing with Stratified 10-fold Cross Validation. Aiming to obtain the best accuracy model as an early warning system for companies, the best model obtained in this study is the one with a kernel size of 3, learning rate α of 0.0001, and L2 regularization on 3 selected layers with an accuracy of 92.31%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	artificial neural network, deep learning, financial distress, one-dimensional convolutional neural network, sektor pertambangan, mining sector
Subjects:	H Social Sciences > HB Economic Theory H Social Sciences > HG Finance > HG4012 Mathematical models Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Marcia Nayfa Fahira Zulfikar
Date Deposited:	09 Aug 2024 03:58
Last Modified:	09 Aug 2024 03:58
URI:	http://repository.its.ac.id/id/eprint/115000

Actions (login required)

View Item