A study about the performance of time series models for the analysis of agricultural prices

Abstract

Agribusiness is one of the most relevant achievements in Brazilian and world economies. Price is an important variable within this sector since economic and market conditions vary over time. Efficient modeling methods are needed to describe better the trends and the characteristics of that variable. The performance of time series techniques of fixed and open models for analysis of agricultural prices will be provided to demonstrate the feasibility of techniques for the generation of results in the wake of economic decisions. Historical time series of monthly average price of cash crops, such as groundnut, sugarcane, banana and orange, received by Brazilian producers, were analyzed for the period between August 1994 and December 2009 and updated by the General Price Index - Internal Availability. Current study shows that exponential smoothing models and Box & Jenkins modeling are viable alternatives for the adjustment of the above price series, each one with its specific characteristics. Keywords: Statistical Models; Agricultural Marketing; Resource Efficiency; Management; Decision Taking.