Research On The Optimal Methods Of The Listed Company Financial Distress Prediction

With the rapid development of Chinese market economy, securities market becomes an important party of our national economy. Although the number of listed companies is increasing and scale is extending continuously, but the unsuccessful financial performance and weak anti-risk capability are still the problems in the most of listed companies. It is of great practical significance for the enterprise, investors and the government how to establish an effective financial distress prediction systems using information technology.The financial crisis of listed companies and data mining concepts are described, and the optimization of financial distress prediction system has been studied in this thesis, including the input dimensions optimization of neural network using rough-set-based reduction technique, and neural network weights and thresholds optimization using the genetic algorithm, and the results of comparing two optimized models with the conventional BP neural network model. The concrete contents including:(1) rough-set-based reduction technique are used for early-warning index reductions, so that reduce complexities of neural networks and improve network speed and prediction accuracy;(2) genetic algorithm are used as the pre-installation of neural network model to optimize the network input value and the threshold, so as to short the network training time and improve the prediction accuracy. Empirical studies have shown that the prediction of the optimized model is more accuracy than traditional BP neural network model.Based on analysis, modeling and mining on the financial data of listed companies in the past three years, the establishment of financial distress prediction system can work properly,and help enterprises to find the crisis.It also takes measure to keep away from potential crises, improve the financial crisis early-warning capability.