Integrated Evaluation of Listed Companies by Factor Analysis for Symbolic Data

Symbolic data analysis is a new data mining technology. Interval number is a most important type of symbolic data. An interval number can be seen as an ordered pair composed of its center and radius, where the radius can be considered as its limit error. A factor analysis is firstly performed on the center sample data matrix, from which the center factor scores are obtained. The limit errors of the factor scores are then obtained by the radius sample data matrix based on the error transferring formula. As a result, the interval factor scores are derived through the centers and limit errors of the factor scores. Accordingly the integrated behavior of the listed companies is evaluated from the interval factor scores. An empirical research on stocks´ transaction data of twenty listed companies in a certain week of Shanghai financial market is performed. The twenty listed companies are classified into four groups according to their integrated behavior in the market.