Analysis and Estimation of Castor Bean Crop Productivity

This study proposes a statistical model to estimate the productivity (kg/ha) of castor beans in the state of Cear? as a function of rainfall throughout their vegetative and reproductive cycle. Seven municipalities located in the region known as "Sert緌 Central" were selected; this choice is justified by the fact that this is the most traditional region for castor bean cultivation in the state of Cear? Correlation coefficients were calculated between productivity and various parameters related to precipitation (total rainfall, rainfall at the beginning and end of the castor bean cycle, duration of the rainy season, etc.). To minimize the effect of local factors, such as management and soil type, the precipitation and productivity variables were normalized. Next, to eliminate the redundancy in the original variables, principal component analysis (PCA) was used, which reduced the number of independent variables from six to two. By performing a multiple regression of productivity with these two new variables, a model was obtained that allowed the estimation of the harvest for the region studied.