Uso de modelos de séries temporais para previsões de safras de milho no estado de Mato Grosso

Abstract

The Mato Grosso State is the main producer of corn of the Brazil and its production has been increasing every year. In this sense, is very important to gain information about future production to planning and monitoring of the corn crops. In this way, the main aim of this paper is to compare the performance showed by the forecast models of time series and to choose the best model. The historical data of corn crop from 1976/1977 to 2017/2018 was obtained with CONAB (The Brazilian National Supply Company). Then, the time series pattern was analyzed, as well as the descriptive statistics of the data obtained. Subsequently, electronic spreadsheets were developed for application and analysis of the evaluated models. With the results it was verified that the trend exponential smoothing model (Holt's linear model) presented the smallest prediction errors, and then it was selected to predict the next seven crops (from 2018/2019 to 2024/2025). The forecast obtained by this model for the 2024/2025 crop indicates that total corn production in the state of Mato Grosso will increase by approximately 70% compared to the 2017/2018 crop production.

Previsões De Indicadores Da Cultura Do Algodão No Estado De Mato Grosso

“…Logo, realizar previsões dos indicadores da cultura do algodão do estado que é o principal produtor se configura como uma atividade muito importante. Segundo Silva et al (2019), prever indicadores de culturas agrícolas é essencial para o planejamento de safras e para o processo de tomada de decisões em geral.…”

Techniques of quality of adjustment of statistical models with evaluation of probability distributions using production data of laying quails

“…Making it a useful tool in reducing errors, costs, and processing time. Like other authors, they used statistical modeling techniques to forecast corn crops in the state of Mato Grosso, as they realized the need to predict events such as rain or drought, based on past situations (Silva et al, 2019).…”