A Hybrid Model to Identify the Financially Distressed Firms

Khalid Mumtaz Khan

Abstract


This paper aims at identifying the financially distressed firms in the housing sector in US, well before their actual bankruptcy occurs, using a composite model by integrating accounting model and option-based model. Housing sector in the US acts as providing the baseline assets for securitizing by creating prime mortgage instruments, and further sub-prime mortgage instruments. Most of the bankruptcy prediction studies exclude housing sector from their sample due to its distinct characteristics. Bankruptcy prediction models mostly use accounting input, diluting the fact that accounting data is tailored for financial reporting rather than bankruptcy prediction. It has innate limitations to be used as an input for bankruptcy prediction models. In the recent past several studies proposed using option-based models for bankruptcy prediction. These models assume of stock markets being efficient, which may not be true in the real life. This paper proposes a hybrid model composed of accounting model and option-based model to identify the financially distressed firms, in time. Area Under the receiver operating characteristic Curve AUC is used as the indicator of the predicting performance of the proposed hybrid model. Empirical results show that the proposed model has the highest AUC. This result show that neither accounting information-based models, nor the stock market information models alone is sufficient to explain the real-world pattern of financial distress leading to bankruptcy. A hybrid model composed of both accounting, and stock market information is better predictor of bankruptcy. It is proposed that the propose hybrid model be used to identify financially distressed firm in housing sector in the US.

Key words:      financial distress, bankruptcy, accounting model, option model, hybrid model, US housing sector.


Refbacks

  • There are currently no refbacks.


............................................................................................................................................................................................................................

HOW DO YOU REGISTER and SUBMIT AN ARTICLE?

Registering and Logging in

Submitting an Article