Search results

The purpose of this paper is to identify if peer firms’ capital structure decision plays a role in determining focal firms’ capital structure decision, despite the fact that correlated effects can also lead to co-movement of financing behavior among firms from the same industry (i.e. industry-specific capital structure).

Instead of using relative measurement (of individual outcome variable over industry variable) as in previous work, this paper borrows the linear-in-means model and, after controlling for potential endogeneity problems, directly identifies the existence of peer effects with coefficient estimation. To deal with correlated effects, additional empirical investigations such as test of heterogeneity in direction and scale, social multiplier identification test and instrumental regression test based on another instrumental variable (that is less influenced by correlated effects) are performed.

Using data from Chinese listed firms, this paper, for the first time, identifies the presence of peer effects in capital structure and debt maturity decision. Further investigations show that first, focal firms react asymmetrically to peer firms’ debt adjustment of different direction and scale. Second, social multiplier, a unique attribute of peer effects, is identified in the leverage choices. Third, the significant correlation of capital structure decision remains even if we use another “correlated effects-immune” instrument. All these results point to the fact that peer effects, rather than correlated effects, play a significant role in determining capital structure.

The empirical results of this paper provide strong evidence that firms, driven by motivations such as either learning or competition, will actively react to peers’ financial decisions. As the bridge between individual firms and the industry, social multiplier can be fully taken advantage of to induce positive spillover of good management practices and prohibit inefficient decisions from spreading.

This paper theoretically and empirically introduces peer effects – a well-acknowledged social concept – into capital structure decision of Chinese listed firms, thus both complementing the traditional capital structure theory and providing an empirical paradigm for peer effects research.

One of the main strengths of qualitative research is to focus on ideas, concepts and meanings involving individuals and groups of people in their own discourse. Instead of testing a narrow hypothesis or making a generalisation about a population on certain issues under investigation, qualitative research attempts to present different insights which can only be unearthed by direct and personal engagement with research participants (Brannen, 1992). This engagement should take place in a natural social context where real life takes place. However, conducting qualitative research in China can pose a huge challenge for both Chinese and international researchers. This paper examines some problems (ethics, linguistics, etc) of using qualitative research methods and tools such as interviews, participant observation, and Critical Discourse Analysis in China.

In recent years, the demand for road traffic has continued to increase, but the casualties and economic losses caused by traffic accidents have also remained high. Therefore, the use of social service robots to manage, supervise and warn real-time traffic information has become an inevitable trend of traffic safety management.

In order to explore the inherent objective development law of road traffic accidents, in this paper, the factor analysis (FA) is used to explore the main influencing factors of traffic accidents, then the random forest algorithm is applied to build an FA–RF-based road traffic accident severity prediction model to predict two- and three-category accidents.

By comprehensively comparing the classification results of the two- and the three-category accident prediction, it also finds that due to the intersection between injuries and fatalities and the lack of necessarily external environmental information, the FA–RF model has a large degree of misjudgment for injuries and fatalities. Therefore, it is recommended to establish a real-time autonomous information communication mechanism between different kinds of social robots, which can improve the prediction of traffic accidents.

(1) A fusion model of FA–RF is considered to predict traffic accidents, which can be applied in traffic service robot. (2) It is recommended to establish a real-time autonomous information communication mechanism between different kinds of social robots, which can improve the prediction of traffic accidents.