Identifying Influencers of Corporate Performance in Inter-firm Networks
2017 Portland International Conference on Management of Engineering and Technology (PICMET)
The recent volatility of the business environment has increased the importance of firms that could foster innovation and influence other firms to bring competitive advantage to their surrounding regions. Therefore, it is becoming increasingly important for policymakers to identify such firms to maximize the effects of subsidies aimed at stimulating the regional economy. The purpose of this research is to build practical prediction models of influencers of corporate performance by extracting features of firms that have an impact on surrounding firms' corporate performances. Firms are defined as influencers of corporate performance when an increase in the revenue of surrounding firms was observed in the fiscal year after their own revenue increased. Inter-firm relationships of firms located in Aichi prefecture that Teikoku Databank, Ltd. possesses is used as the target of this research. Inter-firm networks are formed to calculate network indices such as network centralities and participation coefficient. Those network indices are used as features for building prediction models classifying the influencers. As a result, the f score over 0.7 was realized. Network index such as betweenness centrality contributed significantly on the prediction models. This paper contributes to explaining the implication of the index in the propagation of corporate performance within an inter-firm network.
＞For more details, please refer to the link here.
The University of Tokyo, Sakata laboratory, Taro Oka, Hajime Sasaki, and Ichiro Sakata