Skip to main content

Table 5 Loan default classification comparison of credit and profit scoring models—lending club

From: Default or profit scoring credit systems? Evidence from European and US peer-to-peer lending markets

Model Accuracy (%) Specificity (%) Sensitivity (%)
Panel A: Credit scoring models    
 LR 67.5 67.2 67.6
 \(LR^{{\lambda_{\min } ,\alpha = 1}}\) 67.5 66.7 67.7
 \(LR^{{\lambda_{\min } ,\alpha = 0}}\) 67.9 66.7 68.1
 \(LR^{{\lambda_{\min } ,\alpha_{\min } }}\) 67.6 66.4 67.9
 RFC 75.7 42.6 83.7
 NNC 72.1 55.1 76.2
Panel B: Profit scoring models    
 LM 69.1 51.3 73.4
 \(LM^{{\lambda_{\min } ,\alpha = 1}}\) 72.4 49.7 77.9
 \(LM^{{\lambda_{\min } ,\alpha = 0}}\) 71.4 47.9 77.0
 \(LM^{{\lambda_{\min } ,\alpha_{\min } }}\) 72.6 49.5 78.1
 RFR 72.0 46.4 78.1
 NNR 73.8 33.8 83.5
  1. \(LR^{{\lambda_{\min } ,\alpha = 1}}\), \(LR^{{\lambda_{\min } ,\alpha = 0}}\), \(LR^{{\lambda_{\min } ,\alpha_{\min } }}\) are lasso, ridge and elastic net version of logistic regression, \(LM^{{\lambda_{\min } ,\alpha = 1}}\), \(LM^{{\lambda_{\min } ,\alpha = 0}}\), \(LM^{{\lambda_{\min } ,\alpha_{\min } }}\) are lasso, ridge and elastic net version of linear regression
  2. RFC random forest classification, NNC neural network classification, RRR random forest regression, NNR neural network regression.