Initial Values in OSR

Dear all,

I have a question about using the command “osr” in a simulation. It seems that different initial values provide me with quite different optimal parameters in a simple rule. Is this an indicator of any issues with my model? How should I choose initial values?

Thanks a lot in advance.

It just indicates that your target function has several local minima and your local optimizer gets stuck. Experiment with different starting values and choose the global minimum.