African Journal of
Business Management

  • Abbreviation: Afr. J. Bus. Manage.
  • Language: English
  • ISSN: 1993-8233
  • DOI: 10.5897/AJBM
  • Start Year: 2007
  • Published Articles: 4203

Full Length Research Paper

Predicting corporate distress in the Nigerian stock market: Neural network versus multiple discriminant analysis

Peter Omohezuaun Eriki
  • Peter Omohezuaun Eriki
  • Department of Banking and Finance, Faculty of Management Sciences, University of Benin, Benin City, Nigeria.
  • Google Scholar
Ralph Udegbunam
  • Ralph Udegbunam
  • Department of Economics and Statistics,University of Benin, Benin City, Nigeria.
  • Google Scholar


  •  Accepted: 04 October 2013
  •  Published: 14 October 2013

Abstract

The objective of the paper is to assess the quality of neural networks in predicting distress as against discriminant analysis and its applications in enhancing managers’ decision. Forty four firms listed on the Nigerian Stock Market between 1987 and 2006 are used for the study.  The performance of neural network is then compared with the more familiar discriminant analysis statistical technique, and the performance of both is further with a performance obtainable by mere guesswork. The results show that, while both the neural network and the discriminant analysis techniques performed better than guess work, the neural network out performs the discriminant analysis technique. The outstanding performance of neural network underscores its importance as an invaluable tool in the business decision-making process. The study suggests that neural networks could aid managers in decision making to reverse the present down trend of the Nigerian Stock Market.

Key words: Stock market, neural networks, firm’s distress, business decision, pay-out policy, discriminant analysis.