Discovery of abnormal financial behavior in a dynamic finance environment with hierarchical self-organizing mapping

That many domestic corporations have suffered from financial crises recently victimizes shareholders of those corporations and. Prior researches about those financial crises are basically one-dimensional basing on static balance sheets or variables of macroeconomics. The present research is directed to utilizing self-organizing mapping (SOM) of a neural network in the dynamic environment for discovering the abnormal financial behaviors of the corporations. SOM, which has been applied by a number of scholars previously in their analyses of detecting the financial crises, is for obtaining the rules for internal clusters of the input data before categorizing them. The present research relies on the macroeconomics and financial indicators, first-order variable processing, normalization, and utilization of the hierarchical SOM to establish the trend of the financial behaviors of the corporations. The present research further shows the combination of macroeconomics indicators and financial indicators is superior to the use of financial indicators only in predicting the abnormal corporation financial behaviors. And the utilization of the hierarchical SOM helps identify the abnormal financial behaviors associated with the corporate operations more effectively.