Research Article

Financial Distress Prediction of Chinese Listed Companies Using the Combination of Optimization Model and Convolutional Neural Network

Table 9

Prediction results of different models.

YearsCriterionNNCARTSVMLROurs

2012–2016Accuracy92.66%80.39%92.37%88.59%94.5%
KS0.81640.59870.80870.7380.9857
AUC0.85770.61260.85170.77860.9974

2012–2015Accuracy72.98%54.38%86.90%85.10%88.52%
KS0.37710.50840.64150.60250.7774
AUC0.5130.29080.75390.72340.9548

2016Accuracy83.05%82.15%91.09%88.92%N/A
KS0.57690.62660.76890.7224N/A
AUC0.68560.65210.82860.7881N/A

2015Accuracy85.52%77.92%84.09%86.38%N/A
KS0.59060.54010.60160.6526N/A
AUC0.7310.5690.70250.743N/A

2014Accuracy68.82%70.17%76.12%73.54%N/A
KS0.34510.3870.29830.2884N/A
AUC0.46580.43520.57930.5383N/A

2013Accuracy76.38%76.12%79.42%71.48%N/A
KS0.38270.46030.38980.2662N/A
AUC0.57890.56050.63060.508N/A

2012Accuracy72.46%71.67%72.19%64.71%N/A
KS0.33460.38810.30250.3435N/A
AUC0.51960.47990.52010.4175N/A