Neural Network Simulation and the Prediction of Corporate Outcomes: Some Empirical Findings

Nicholas Wilson, Kwee Chong, Michael Peel and A. N. Kolmogorov

International Journal of the Economics of Business, 1995, vol. 2, issue 1, 31-50

Abstract: Neural Networks (NN's) involve an innovative method of simulating and analysing complex and constantly changing systems of relationships. Originally developed to mimic the neural architecture and functioning of the human brain, NN techniques have recently been applied to many different business fields and have demonstrated a capability to solve complex problems. This paper investigates the use of NN techniques as a tool for the modelling and prediction of corporate bankruptcy and other corporate outcomes. The within and out-of-sample accuracy of trained NNs are compared with those of standard logit and multilogit techniques. The results of the study suggest that, from a pure predictive point of view, NN simulation produces a higher predictive accuracy and is more robust than conventional logit and multilogit models.

Keywords: Corporate failure; Neural networks; Multi-outcome models, JEL classifications: G33, G34, C53, (search for similar items in EconPapers)
Date: 1995
References: View complete reference list from CitEc
Citations: View citations in EconPapers (11)

Downloads: (external link)
http://www.tandfonline.com/10.1080/758521095 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:ijecbs:v:2:y:1995:i:1:p:31-50

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/CIJB20

DOI: 10.1080/758521095

Access Statistics for this article

International Journal of the Economics of Business is currently edited by Eleanor Morgan

More articles in International Journal of the Economics of Business from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().