Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Corporate failure for manufacturing industries using financial ratios and macroeconomic variables with logit analysis
Abstract
This study utilized both financial and market information in the prediction of corporate failure. Six financial ratios and three market variables based on a theoretical justification were used. A sample of 110 manufacturing companies which had become bankrupt between 1980 and 1987 were identified from the F & S Index and matched to 110 non-failed companies on the basis of total assets, financial statement date, and four digit industry code. Financial data were collected for a six year period from COMPUSTAT tapes while market data were collected from The Economic Report of the President, Business Condition Digest, and Standard and Poors' Stock Price Index. Using logit analysis, the following four models for prediction of corporate failure were developed: (1) a model which used financial ratios alone; (2) a model which used financial ratios with changes in macroeconomic variables; (3) a model which used changes in financial ratios; and (4) a model which used changes in financial ratios with changes in macroeconomic variables. Of the four models estimated, model 2 (financial ratios and changes in macroeconomic variables), which had the largest explanatory power as measured by its likelihood ratio index, was employed in all further analysis. This model predicted correctly 87.82 and 87.50 percent of the estimation and holdout samples, respectively. Given these results, this model compared favorably with other failure prediction models in the literature. The significance of the coefficients in each year's model was evaluated by using the t-statistic corresponding to each coefficient's value. The accuracy of the selected model was also evaluated on the basis of misclassification costs for different costs of type I and type II errors when the model was applied to an estimation and a holdout sample. The model misclassification costs are lower than the misclassification costs of a naive model. The results indicated that inclusion of market variables does improve predictive accuracy of the models.
Subject Area
Accounting
Recommended Citation
Al-Darayseh, Musa M, "Corporate failure for manufacturing industries using financial ratios and macroeconomic variables with logit analysis" (1990). ETD collection for University of Nebraska-Lincoln. AAI9030100.
https://digitalcommons.unl.edu/dissertations/AAI9030100