Search

From the network perspective, this paper explores the effect of FDI on trade and analyzes the structural effect on different sectors classified by technology levels among the Belt and Road (B&R) countries. Employing social network analysis methods, we depict the evolution of FDI networks and trade networks of B&R countries and find that both two networks are disassortative and small-world networks as well as the density and number of edges increase. Through the results of ERGM, we get three findings about the effect of FDI on trade. (1) FDI has the positive effect on trade among B&R countries from the network perspective; (2) the positive effect of FDI on trade has improved since the Belt and Road Initiative is proposed; (3) FDI promotes the exports of low technological sector, medium technological sector, and high technological sector, but the effect on medium technological sector is stronger than the other two sectors. [ABSTRACT FROM AUTHOR]

Exponential family random graph models (ERGM) are increasingly used in the study of social networks. These models are build to explain the global structure of a network while allowing inference on tie prediction on a micro level. The number of papers within economics is however limited. Possible applications for economics are however abundant. The aim of this document is to provide an explanation of the basic mechanics behind the models and provide a sample code (using R and the packages statnet and ERGM) to operationalize and interpret results and analyse goodness of fit. After reading this paper the reader should be able to start their own analysis. [ABSTRACT FROM AUTHOR]