To make high-quality research more accessible and easier to explore.

Fields:
2 results

Recession‐Induced Stress and the Prediction of Corporate Failure*

Contemporary Accounting Research 1996 13(2), 631-650
In this paper we examine whether the occurrence of recession‐induced stress is an incrementally informative factor that contributes to the predictive and explanatory power of accounting‐based failure prediction models. We show that accounting‐based statistical models used to predict corporate failure are sensitive to the occurrence of a recession. Moreover, after controlling for the intertemporally unconditioned “stressed” and “unstressed” types of corporate failure, we find that models conditioned on the occurrence of a recession still add incremental explanatory power in predicting the likelihood of corporate failure. This source‐related characterization of stress appears distinct from other types of corporate failure that have been identified. Résumé. Les auteurs se demandent si l'occurrence du stress amené par la récession est un facteur qui apporte une information supplémentaire contribuant au pouvoir prédictif et explicatif des modèles de prévision des faillites reposant sur la comptabilité. Ils montrent que les modèles statistiques fondés sur la comptabilité utilisés pour prévoir les faillites des entreprises sont sensibles à l'occurrence d'une récession. De plus, une fois contrôlée la nature de la faillite de l'entreprise — faillite annoncée par le stress et faillite non annoncée par le stress sans conditionnement intertemporel —, les auteurs en viennent à la conclusion que les modèles conditionnés par l'occurrence d'une récession ont encore un pouvoir explicatif accru dans la prédiction de la probabilité de faillite de l'entreprise. Cette définition du stress liée à la source semble différente des autres types de faillite de l'entreprise qui ont été cernés.

Rank Transformations and the Prediction of Corporate Failure*

Contemporary Accounting Research 1998 15(2), 145-166
Rank transformation of observations has been shown to be useful in linear modeling because the models so constructed are less sensitive to outliers and/or non‐normal distributions than are models constructed using standard methods. In the present study, we apply rank transformations to financial ratios to improve the predictive usefulness of standard failure prediction models. Kane, Richardson, and Graybeal (1996) have shown that failure prediction can be improved by conditioning accounting‐based statistical models on the occurrence of recession. Our results suggest that rank‐ transformed data models show additional improvement in prediction without the added cost of having to predict recession for the companies undergoing testing for potential failure.