Research on Financial Warning for Listed Companies by Using BP Neural Networks and Panel Data
Leping Wang · Systems Engineering - Theory & Practice · 2007
On the basis of the authors' previous research and with the introduction of panel data,this paper constructed a BP neural networks model to predict the financial status of listed companies,by taking panel data composed of the financial data of T-2 and T-3 as the sample.The research indicates that: 1) with panel data,the BP neural networks warning analysis is of stable and continuous predictability,which is suitable in constructing practical mid-term and long-erm prediction models to make the model more applicable;2) the predictability precision is 88.46% and 75.64% for T-3 and T-4,respectively,superior to that of counterparts and the previous research.