Previsões estatísticas e estudo do cenário agrícola e econômico do trigo para o estado do paraná

Abstract

The present study aimed to evaluate and predict the planted areas, harvested areas and prices of wheat grain in the state of Paraná, which is currently the largest producer of this raw material in Brazil. For the beginning of the work, a bibliographical study was presented to present the scenario of the wheat culture in Brazil, the concepts and the stages of a prediction model and concepts about measures and controls of forecast error. After that, historical data were collected directly from the database of the Center for Advanced Studies in Applied Economics (CEPEA) and the Brazilian Institute of Geography and Statistics (IBGE). These data were statistically analyzed to identify patterns, cycles, seasonality and trend. With the help of the data analysis, we chose the forecast model that best fit for each series of data analyzed. The evaluation and treatment of the data occurred through Exponential Smoothing models, through the software Action Stat®. The models found for the prices practiced were multiplicative level, multiplicative trend and seasonality absent. Already the models of planted area and harvested area were multiplicative level and tendency and seasonality were absent. The predictions found for all variables were satisfactory, since they were within the upper and lower limits of the forecast.