Comprehensive Evaluation of Competitiveness of Listed Companies Using Artificial Neural Network

Based on the annual report data between 1995 and 2005 of all listed companies (LCs), the 25 initial financial indexes, widely used by experts and researchers aboard and at home, was deduced to 14 effective evaluation indicators using factor analysis and principal component analysis (PCA). The 14 evaluation indicators, covering five aspects for comprehensive evaluation of competitiveness of LC, retain 85% variance of the 25 initial indexes. An evaluation criteria system with 14 indicators for evaluating the competitiveness of LCs was thus put forward in this paper. Artificial neural networks (ANNs) were used to comprehensively evaluate the competitiveness of LC. The training set data, verification set data and testing set data were generated according to the evaluation criteria system for training, monitoring the training process on-line and judging the performance of the ANNs model. The verification set data was used to monitor the training process to escape from the over-training and the global minimum was reached eventually. The eleven LCs mainly related to communication and culture industries were comprehensively evaluated as examples. The case study results show that the criteria are reasonable and the process of establishing the ANNs model is reliable, feasible and proper.