Identifying companies' bankruptcy using an enhanced neural network model: a case study evaluating the bankruptcy of Iranian stock exchange companies
by Shahin Ordikhani; Sara Habibi; Ahmad Reza Haghighi
International Journal of Industrial and Systems Engineering (IJISE), Vol. 38, No. 4, 2021

Abstract: The purpose of this research is to utilise an enhanced neural network model to anticipate the bankruptcy of stock exchange companies and test the predictive power of this model by considering the concept of misclassification. Misclassifications decrease the accuracy of prediction. Among every one of the structures of the two-layer neural network, the perceptron model with the structure of nine neurons in the input layer and a neuron in the output layer with the Levenberg learning algorithm demonstrated the most predictive power. The neuron structures three, five, and nine were considered to decide the proper characteristics of a two-layer perceptron for anticipating companies' bankruptcy. Among them, the two-layer perceptron with nine neurons in the input layer and one neuron in the output layer identified to has the best performance. The findings demonstrate that applying artificial neural network models amplify financial management for facing with fluctuations and bankruptcy.

Online publication date: Mon, 09-Aug-2021

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Industrial and Systems Engineering (IJISE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com