The study of a financial crisis prediction model based on XBRL

Recent outbreak of corporate financial crises worldwide has brought attention to the need for a new international financial architecture which rests on crisis prediction and crisis management. Financial data have been widely used by researchers to predict financial crisis, but few studies exploit the use of non-financial indicators in corporate governance to construct financial crisis prediction model. This article introduces a prediction model based on a relatively new machine learning technique, support vector machines (SVM) with XBRL financial reporting. This study indicates that the prediction model considering both financial and non-financial information outperforms those models based on only one type of information. Two well-known prediction models, regression model and genetic algorithm, are compared with SVM. The experiment results show that the combined use of both financial and non-financial features with SVM model leads to a more accurate prediction of financial distress.