The wheat is a product used in the manufacture of several products. The industries, when effecting the purchase of this cereal, observe some parameters of industrial quality, that can be determined through some tests (weight hectolítro (PH), gluten strength, Falling Number), carried out in specialized laboratories. In Brazil, the main producers of wheat are the states of Rio Grande do Sul, Paraná and Santa Catarina, and in these states, the climatic variability is great and this can be a factor of influence in the results of the quality tests. This work research is aimed at making a statistics evaluation, of some meteorological variables that affected the industrial quality of the wheat in some regions of the state of Paraná in the years 1999, 2000 and 2001. This evaluation will be carried out through techniques of the multivaried analysis. Thus, a model of linear regression multivaried was adjusted to try itself to explain the relationship of the set of changeable answers IQI (of the quality) with the set of explicativas variables IQP. In the same way it was esteem canonic correlation enters these two sets of variables, that express the relationship between the two sets. The models of multivaried regression developed weren’t enough efficient to predict the quality concerning the climatic variable, however the canonic correlations estimates proved the existence of the relationship between the variables (IQI) of quality and the climatic variables(IQP).
climate; multivariate analysis; canonical correlation
