A Study of Integrating Support-Vector-Machine (SVM) Model and Market-based Model in Predicting Taiwan Construction Contractor Default

Abstract

The construction industry is a high debt ratio, high operating risk and high financial leverage business. The financial instability causes a chain reaction among funds transferred among companies, therefore restricting competitiveness within the industry. The industry’s character and accounting principles of firm-years are different from that of other industries. Creating a hypothetical model of a financial crisis within the construction industry is therefore necessary. Application of this model to real scenarios involving relevant parties can help to forecast a financial crisis in the future. This study applied the market-based model, accounting-based model, hybrid models to predict a financial crisis. These models were then compared to find which can best predict a company that will default. Also, in this paper choosing variables for the Hybrid and Accounting-based models can promote their performance. Finally, the best can be selected for predicting stability.