Indústria de máquinas agrícolas no Brasil: um modelo para estimação da demanda de tratores para o triênio 2016–2018

• Ano de defesa: 2016
• Autor(a) principal: Oliveira, Cristiano Dallagassa Gontijo
• Orientador(a): Ozaki, Vitor Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Economia; Box-Jenkins (1970); Modelos univariados e multivariados; Estimação de demanda; Demanda de tratores agrícolas
• Link de acesso: http://hdl.handle.net/10438/15852

The aims of this work was forecast the demand for agricultural tractors in the Brazilian market during the triennium 2016-2018, using for this, techniques of time series econometrics, in this case, univariate models ARIMA and SARIMA and or multivariate models SARIMAX. Justified this research when holding the industry of agricultural machinery in Brazil, given the economic cycles and other external factors to the economic funda-mentals of demand, where it faces many challenges. Among this, demand estimation stands out because exert a strong impact, for example, planning and cost of short and medium term production, inventory levels, in relation to hand materials and suppliers of local labor and consequently in creating value to the shareholders. During literature review it was found several scientific papers address the agribusiness and its various areas, however, they were not found scientific papers published in Brazil that address the demand forecast of agricultural tractors in Brazil, which served as a motivation for add knowledge to the scientific world and value to the Brazilian market. It was concluded after testing with several models that are presented in the text and appendices, the model SARIMA (15, 1, 1) (1, 1, 1) fulfilled the assumptions set out in the specific objectives to choose the model that best fit itself to the data, and then was chosen as the model to forecast the demand for agricultural tractors in Brazil. These results point to a demand for agricultural tractors in Brazil oscillating between 46,000 and 49,000 units per year between the years 2016 and 2018.