csu home : c.l.a.s.s. : econ dept : campusNet

Myong-Hun Chang

• home •  curriculum vitae • research • teaching • contact •


My SSRN Page


  • A Computational Model of Industry Dynamics

    Routledge (Series: Routledge Advances in Experimental and Computable Economics): January, 2015 [Available at Amazon]

    The study reported in this book attempts to explain many well-documented aspects of the evolution of industries over time. It uses an agent-based computational model in which artificial industries are created and grown to maturity in silico. While the firms in the model are assumed to have bounded rationality, they are nevertheless adaptive in the sense that their experience-based R&D efforts allow them to search for improved technologies. Given a technological environment subject to persistent and unexpected external shocks, the computationally-generated industry remains in a perennial state of flux. The main objective of the study is to identify patterns that exist in the movements of firms as the industry evolves over time along the steady state in which the measured behavior of the firms and the industry stochastically fluctuate around steady means. The computational model developed in this book is able to replicate many of the stylized facts from the empirical industrial organization literature, particularly as the facts pertain to the dynamics of firm entry and exit. Furthermore, the model allows examination of cross-industry variations in entry and exit patterns by systematically varying the characteristics of the market and the technological environment within which the computationally-generated industry evolves. The model demonstrates that the computational approach based on boundedly rational agets in a dynamic setting can be useful and effective in carrying out both positive and normative economic analysis.
  • The computational experiments performed in the book are based on the following baseline C++ codes:
    • MagnumSolo.cpp: generation of time series outputs from a single replication
    • MagnumMult.cpp: generation of time series outputs from 500 independent replications
    • The above C++ codes are created using Microsoft Visual Studio Express 2012. The generated outputs (exported from the codes) are then analyzed using Wolfram Mathematica 7.0.
    • If you do not have a C++ compiler, you may view the above codes in the following text files:

Journal Articles

Book Review

  • Review of Game Theory: A Nontechnical Introduction to the Analysis of Strategy (by Roger A. McCain)
    Eastern Economic Journal, 39 (2013), 267-269.
    [PDF copy]

Working Papers

Recent and Upcoming Conference

* Some of the documents on these sites are in .pdf format (Adobe Acrobat) or .nb format (Mathematica Notebook).  In order to view these documents, you need the following viewers:

  • To download a free Adobe Acrobat Reader, click here: 
  • To download a free MathReader, click here: