Chase Coleman and Dr. Kerk Phillips, Economics
My project pertained to a field of macroeconomics known as Dynamic Stochastic General Equilibrium (DSGE) modeling. DSGE modeling is where an economist builds a model based in economic theory and associates a series of mathematic equations that describe this theory with the model. The key element that makes it more than just algebra is the dynamic feature incorporated with the stochastic element. In order to solve the majority of DSGE models a computer is necessary. Following the simulations, one typically compares the statistical moments that your model returns with those of the economy which you are attempting to simulate.
Much of macroeconomic theory has been built around a seminal paper (Kydland and Prescott 1982) which relied upon a specific technological process to achieve the autocorrelation observed in gross domestic product (GDP) and total factor productivity (TFP). This process for a technology shock z was as follow: z’ = ρz + ε (z’ represents the value of z tomorrow) where ε is distributed normally with mean 0 and standard deviation of 1. While this process successfully matches many key statistical moments, it lacks a solid foundation in economic theory. Dr. Phillips attempted to find a process that was better founded in economic theory in an earlier paper (Phillips Wrasse 2006), but was unable to achieve suitable levels of autocorrelation. The goal of this paper is to build a simple model that can achieve the observed levels of autocorrelation.
Our approach to this problem was a very standard set up. We built a model founded in mathematically sound economic theory and simulated our model in a computer. Since there is no closed-form solution to many complicated models, we used a method known as log-linearization to approximate the path of the economy (This method typically has small errors and is frequently used in macroeconomic literature). We then compared our moments with comparable moments from U.S. data. We built a model where there are two types of firms, production and research. The production firms are monopolists and research firms attempt to innovate to obtain a new technology that would make them the new monopolist in their sector. Without going into the math, the fact that these firms are attempting to repeatedly innovate provides a high level of autocorrelation within the economy. Our results almost exactly match the levels of observed autocorrelation out to a lag of two periods. We believe that adding a higher degree of complexity to the model would have allowed us to match more statistical moments, but that wasn’t the goal of this paper.
While we were working on this project, I had an opportunity to present our paper at two different conferences. I presented both at the Western Social Science Association’s 2013 conference and Midwest Macro 2013. We have since submitted our paper to the Journal of Macroeconomic Dynamics and hope to have it published soon. The paper can be found at: https://economics.byu.edu/Documents/Macro%20Lab/Working %20Paper%20Series/BYUMCL2013-03.pdf