Mixed-influence nonlinear growth model
2003
Prajneshu | Kandala, V.M. (Indian Agricultural Statistics Research Institute, New Delhi (India))
Nonlinear mixed-influence growth model, obtained by combining well-known logistic and monomolecular models, is thoroughly studied. As an, illustration proportion of area under high yielding varieties of wheat in the country during post-Green revolution era is modelled through this model. Employing nonlinear estimation procedures, it is found that some parameter estimates did not converge to biologically meaningful values. Subsiquently, one parameter of the model is reparamaterized by using the concept of `Expected-value parameter'. Consequently, the resultant form is successfully fixed to the data. Another advantage is that, by computing Bates and Watts measures of nonlinearity, it is shown that the model-data set combination depicted close-to-linear behaviour. Finally, Wald's test is applied to test the hypothesis of equality of some parameters.
显示更多 [+] 显示较少 [-]