Two-level self-organizing maps for analysis of financial statements

We propose a novel tool based on a hierarchy of two self-organizing maps (SOMs) for analyzing financial statements. The inputs to the first-level SOM are financial indicators derived from a company´s annual financial statements; these determine the company´s position on the first-level SOM each year. The inputs to the second-level SOM are the coordinates of the company on the first-level SOM during two or more consecutive years. The second-level SOM turns out to give a more accurate description of the state of the company than the first-level SOM. Moreover, it is easy to interpret, as each point on the second-level SOM corresponds to a trajectory on the first-level SOM. With our method, several different patterns of corporate behavior can be recognized