A neural network approach to the prediction of going concern status

Research into the usefulness of a neural network model for the prediction of going concern status, in a company, from financial ratios found it compared well with both the probit model and the auditor's findings. It was also noted the neural network could work better with the inclusion of more information, such as the training data set. There has been some criticism over the methods used by auditors to determine the going concern status of a firm. Some observers have suggested the use of statistical methods may help the auditors in their work and make any conclusions seem less subjective.