다변량 판별분석과 로지스틱 회귀분석, 인공신경망 분석을 이용한 호텔 도산 예측

Prediction of Hotel Bankruptcy Using Multivariate Discriminant Analysis, Logistic Regression and Artificial Neural Network

This study estimates the possibilities of hotel bankruptcy using various analytic methods such as multivariate discriminant model, logistic regression model and artificial neural network model. The estimated models suggest that debt-burned hotels with low profit margin and return on common stockholders’ equity are more likely to be candidates of bankruptcy. The analysis of in-model variables, along with a cross-group comparison of financial ratios, suggests that bankrupt hotels have heavily relied on debt to finance fast sales growth without paying proper attention to controlling their operating expenses and financing costs. Author suggests in conclusion that to avoid bankruptcy risk hoteliers are encouraged to adopt a prudent growth strategy accompanied by less debt financing and tighter cost control.