The Comparison of Machine Learning Model to Predict Bankruptcy: Indonesian Stock Exchange Data

Abstract

This study aims to determine the Machine Learning Model used to predict bankruptcy. The data was conducted from the financial statements of two public companies reported by the Indonesia Stock Exchange from 2009 to 2015. This research method uses an analysis feature in which the accounting ratios are used in statistical analysis of financial statements that handle missing values, choose the correlation feature related to class, and dealing with unbalanced datasets. This problem was resolved at the beginning of the pre-processing phase. The training process uses pre-processing results to fit the data with the prediction model. Accuracy is used to measure the performance of the model in predicting bankruptcy. The result is Sequential Minimal Optimization (SMO) with linear kernel function that works best to predict 1 year before bankruptcy with an accuracy of 91.57% and SMO with Radial Basis Function (RBF) works well to predict 2 years before bankruptcy; the accuracy is 93.8%. This study shows the effect of feature selection and normalization process in making correct predictions using the SMO method.

Forecasting in financial accounting with artificial intelligence – A systematic literature review and future research agenda

“…The predictive power of support vector machines was also evaluated by Antunes et al (2017) by comparing them with the logistic regression and Gaussian processes, with the result that Gaussian processes effectively improved the forecast performance. The fact that support vector machines could make more accurate bankruptcy forecasts than neural networks, decision trees, and logistic regression was also confirmed in the study by Rainarli (2019). The superior predictive power of support vector machines was also demonstrated in the study of Sehgal et al (2021), who compared it with neural networks and logistic regression.…”

“…The majority of the articles that are part of our literature review use AI and ML techniques that rely on supervised learning. There are only a few studies that apply algorithms that belong to unsupervised learning (Shi et al, 2009;Ding et al, 2019;Rainarli, 2019;Brown et al, 2020).…”