Principal component analysis of agricultural data on soybean production

Abstract

Agribusiness plays a fundamental role in the Brazilian economy and has significantly contributed to the country's growth. This article aims to analyze agribusiness data using the principal component analysis technique, validating the relevance to the sector and presenting the possibilities the statistical techniques use bring for maximizing soybean production in the field. The selected variables for analysis include: area, dose per hectare, quantidade consumo, área execução, dose real, valor, vazão, peso bruto, impureza e umidade. The principal component analysis technique was applied to reduce data dimensionality, identify underlying patterns, and investigate relationships among the selected variables. The Kaiser criterion was used to validate the suitability of the data for principal component analysis, considering the variables' eigenvalues to determine the number of significant principal components to be retained. Because of that, it is concluded that principal component analysis, besides being an excellent way to save computational resources by minimizing the volume of variables to be analyzed, also adequately explains the dynamic behavior of information, proving the technique's relevance for understanding the peculiarities of soybean production in agriculture and providing valuable insights for strategic decision-making.