engleski

A method for yield estimation

The problem of yield estimation in plants and animal production is considered. The estimation should be done as early and as exactly as possible. For that purpose the well--known asymptotic growth models are used: general exponential model, von Bertalanffy model, logistic and generalized logistic model. In all mentioned methods the model--function y can be written in the following form: y(t ; A, p)=A g(t ; p), where A is the upper asymptote, and p is the parameter vector, which should be estimated on the basis of measurement data (t_i, y_i), i=1, ..., m. Parameter estimations will be done on the basis of the ordinary least squares method. Since the unknown parameter A in the considered models can always be separated, it is possible to apply a special parameter estimation method whose variables separate. The mentioned method of yield estimation is illustrated with several practical examples.

yield estimation ; growth models ; least squares

nije evidentirano