Abstrait

Application of learning vector quantization neural network in the financial failure prediction

Ying Feng, Caiqin Zhao

Effective prediction of financial failures has been of great importance to Chinese listed companies because it can exert a big influence upon financial decisions to be made by investors, creditors and banking officers. For this purpose, neural network method has been introduced, and it has become a hot spot in this domain. LVQ (LearningVector Quantization) neural networkmethod is adopted to set up a prediction model of financial failure in accordance with latest financial data of 14 listed companies. Repeated training and learning of the sample brings LVQ out.Acomparison between LVQ and traditional BP (Back Propagation) has proved that LVQ algorithmhas a higher prediction accuracy, which indicates that LVQ neural network method will enjoy good application prospect in the field of financial failure prediction.