Financial Early-Warning Model Using Logistic Regression with SCAD Penalty

Logistic regression(LR) is a supervised learning algorithm.And it is widely used in financial early-warning models and analysis.But its model has the potential over-fitting problem.Aiming at this problem,the financial early-warning model using LR with smoothly clipped absolute deviation(SCAD) penalty is put forward.The model can well solve the over-fitting problem and perform variable selection and model parameter estimation simultaneously.And the model's explanatory power is improved.Experiments are implemented on the financial data of A-share manufacturing listed companies of the Shanghai and Shenzhen stock markets.Compared with general,L1 norm penalized and L2 norm penalized LR model,the LR model with SCAD penalty has better classification results and stronger economic interpretations.