Starting Values - Nonlinear Models

Fitting a nonlinear regression requires starting values. The values used for the starting values are dependent on the model chosen. Of course, values close to the true parameter value will reduce convergence issues. A poor choice can cause convergence to a local minimum on the function, which can leads to a suboptimal solution. 

Monomolecular model:

Logistic Model:

Mitcherlich model: