Abstract
Indonesian corporations have been borrowing large sums of money from foreign investors in the past decade, such that private debt ratio has reached 49% of Indonesia’s total external debt by the end of 2017. This act of borrowing might improve the borrowing firms’ performance which leads to increase in profit, but in other hand it might result on debt value expansion, due to the exchange rate depreciation trend in Indonesia. This paper employs Support Vector Regression, a machine-learning method, to study the relationship between factors that might affect corporate performance, and compares the results with that of the conventional panel data regression method. The study was done using data from annual financial statements of 189 firms in Indonesia during 2011-2017.
 It is shown that the machine-learning approach discussed in this study gave better accuracy than the previously employed panel data regression method. Both methods generally showed that balance-sheet effect is more dominant in Indonesian corporations, and it is recommended for companies to minimize their foreign debts and imported purchases, and if possible, export more of their products.
Highlights
Indonesia’s external debt has been rising in the past decade, with 10.03% year on year growth rate
The NFA threshold is -3.03%, which means if a firm has more than 3.03% debt in foreign currency relative to its asset, the firm would most likely experience decrease in its equity when depreciation happens
From this study, we conclude that support vector regression could predict a firm’s equity with better accuracy compared to the conventionally used ordinary least squares (OLS) panel regression method
Summary
The Central Bank of Indonesia derived the factors that affect a firm’s performance from their balance sheet. A firm’s performance is evaluated by its equity to asset ratio [2]. At time 0, a company’s balance sheet could be defined as. Index Terms – depreciation; exchange rate; external debt; machine learning; support vector regression
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: European Journal of Electrical Engineering and Computer Science
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
