Applying Grey System Theory to predict default probability in construction industry

The prediction of business failure is an important and challenging issue that has served as the impulsion for many academic studies over the past decades. The widely applied methods to predict the risk of business failure were based on financial ratio analysis; therefore, the purpose of this thesis is to explore the possibility of applying Grey System Theory for predicting default probability of construction firms by analyzing financial history data. In the present research, the author introduces a framework of the bankruptcy probability valuation relying on the 19 initial financial ratios. First, Synthetic Degree Incidences ρ of considered firms are calculated and combine these ρ values, the default probability of firms as well as the importance of financial ratios as predictors will be identified. In this research scope, the author proposes 3 models correspond with difference history consequence data (3 years, 4 years and 5 years). Then, using ROC curves to point out which one is the most favorable consequence data for model (correspond to the highest AUC value). After the suitable consequence data (4 years) was chosen, the other effected factors like as X0, θ and the combo of key variables were considered. Some warnings as well as recommendations are highlighted for those who concern of applying grey analysis model to forecast default probability of construction firms.