An effective going concern prediction model for the sustainability of enterprises and capital market development

The purpose of this study is to construct a two-stage effective and innovative going concern prediction model to predict going concern doubt for the sustainability of enterprises and capital market development. Samples of this study are the companies listed on the Taiwan Stock Exchange or the Taipei Exchange, totalling 196 companies and including 49 companies with going concern doubt and 147 normal companies (with no going concern doubt). The data are taken from the Taiwan Economic Journal (TEJ) and the Market Observation Post System during the period from 2001 to 2016 (totalling 16 years). This study adopts a two-stage way to construct the going concern prediction models. In Stage I, the traditional statistical method of stepwise regression (SR) and the data mining technique artificial neural network (ANN) are applied to select the important variables. In Stage II, two decision tree algorithms (data mining techniques): classification and regression tree (CART) and C5.0 are used to establish the prediction models. The results show that the SR + CART model has the highest going concern prediction accuracy, with an overall accuracy of 87.42%.