African Journal of
Business Management

  • Abbreviation: Afr. J. Bus. Manage.
  • Language: English
  • ISSN: 1993-8233
  • DOI: 10.5897/AJBM
  • Start Year: 2007
  • Published Articles: 4164

Full Length Research Paper

Detecting firms with going-concern risk based on the industry affiliation, corporate governance characteristics, and financial performance

Yuh-Jiuan M. Parng1* and Chung-Jen Fu2
1Department of Accounting and Information System, Asia University, Taichung, Taiwan, R.O.C. 2Department of Accounting, National Yunlin University of Science and Technology, Douliu, Yunlin, Taiwan, R.O.C.  
Email: [email protected]

  •  Accepted: 28 June 2011
  •  Published: 30 September 2011


Detecting firms with going-concern risk is precisely critical to all financial professionals. The analytical features included three aspects: the industry domain, the corporate governance characteristics, and the financial performance. An enhanced two-step analytical approach was developed in this study. First, the multivariate analysis (MA) applied to explore influential factors affected the uncertain behaviors of a firm. Secondly, with the prioritized significant factors identified in MA model, the classification and regression tree (CART) technique was adopted to generate decision tree. There were nine significant factors: size of the board of directors, percentage of independent directors, ratio of shares pledged, family-owned type, ratio of cash right deviation, hiring Big 4 CPA firms, earnings per share, debt ratio, and return on assets. These practical finding provides comprehensive understandings of the behaviors of the firms with C-G risks. Weighing from the decision tree modeling, the testing results showed 87.5% successful rate which demonstrated itself as an effective and analytical tool and will suffice the practical needs for detecting firms with going-concern risk. 


Key words: Going-concern, industry affiliation, corporate governance, financial performance, multivariate analysis, classification and regression tree (CART), decision tree.