Article Preview
Top1. Introduction
The past decades have witnessed an increasing interest in studying a nation’s product pace structure and economy development. Among these contributions, Hidalgo et al. introduce an outcome-based measurement to quantify proximities among the products that countries export, and use network representation to visualize proximities among products (Hidalgo, Klinger, Barabasi & Hausmann, 2007). Through which, they apply theory and methods from physics and economics to model and map the impact of the product space on the development of nations (Hidalgo, Klinger, Barabasi & Hausmann, 2007). They find that more sophisticated products are located in a densely connected core, whereas less sophisticated products occupy a rather sparse periphery (Hidalgo, Klinger, Barabasi & Hausmann, 2007). They also find that the product space structure governs the evolution of comparative advantage of nations (Hausmann & Klinger, 2006; Hausmann & Klinger, 2007), and the dynamics of countries’ productive structures is characterized by a few highly dynamic while the product space remains relatively stable (Hidalgo, 2009). Further, they study the characteristics of the relationship between products and the countries, and find that countries differ not just in how diversified they are, but also in the ubiquity of the products they export (Hausmann & Hidalgo, 2010). Using the concepts underlying the construction of product space, Abdon and Felipe study the evolution of productive structure of exports of the Sub-Saharan region, and analyze the opportunities for growth and diversification (Abdon & Felipe 2011). Besides, Hamwey et al. propose a green product space methodology to map the export strengths of countries for a specified set of green products based on the product space model and find that countries can more quickly and accurately identify green sectors through product space analysis (Hamwey, Pacini & Assunção, 2013).
Following their work, this paper aims to estimate the economic agglomeration and product space structure evolutions of a nation by using emergent approaches. Product space evolution involves complex nonlinear interactions and self-organized processes among many enterprises or firms, agent models built on such concepts have been therefore an appropriate tool to study the problems. Among agent simulation models, an agent is defined either complex or minimalistic ways. A complex agent can be regarded as an autonomous entity with either knowledge or behavior based rules, performing complex actions such as learning and building its own strategy with multiple attributes (Muller, Wooldridge & Jennings, 1997). The conceptual design of complex agent is ideal but impractical. The alternative is the minimalistic agent, which has the simplest rule set to guide its decision, without referring the internal attributes. But due to oversimplification, the practical application of such agent is also very limited. To avoid both extremes, Brownian agent approach is proposed (Ebeling, Schweitzer & Tilch, 1999; Schweitzer, 1998; Schweitzer, 2003). A Brownian agent is a minimalistic agent with internal degrees of freedom. Through specific action, Brownian agents are able to “generate a self-consistent field which in turn influences their further movement and behavior” (Schweitzer, 1998). Brownian agent model has been widely applied into many research fields, such as biology (Ebeling & Schweitzer, 2003; Mach & Schweitzer, 2007; Birbaumer & Schweitzer, 2011; Garcia, Birbaumer, & Schweitzer, 2011), sociology (Schweitzer, 2004; Schweitzer & Garcia, 2010; Lee, Park & Moon, 2014), and information technology (Shin & Park, 2009; Sobkowicz & Sobkowicz2010).