Analysis of the economic cycles of Mexico through artificial neural network modeling: 1980 - 2007

Abstract

The knowledge of business cycles is vital for government and financial institutions of a nation, as the management of a healthy economy is associated with making financial and economic decisions based on timely and accurate information. The objective of this paper is to analyze the business cycles in Mexico from 1980 to 2007 using artificial neural networks as techniques to get a better definition, the effect of macroeconomic variables that affect the country’s economy and its economic cycles, thus determining also the different phases that conform it. Fifty one macroeconomic variables directly involved in the behavior of the Mexican economy were analyzed to determine: the number of economic cycles that the country has experienced in the last years, the duration and intensity of the cycles, the speed to reach the critical and maximum points of the economy and the structure of the economy’s phases per cycle. This analysis aims to provide a tool to determine the behavior of the phenomenon of business cycles by integrating a comprehensive approach to the major macroeconomic variables and to have elements of public policy, production, fiscal, monetary and trade among others.