OSR Optimal Simple Rules

Dear Community,

I am trying to use the OSR command to work out the optimal policy parameters in the Taylor rule.

However I just realised that the results change by changing the initial values of the parameters.
In the model attached for example the starting values are the usual (phipi = 1,5/ phiy = 0.5) and the algorithm picks as optimal values phipi = 26.25/ phiy = 1.62. In any way I change the initial conditions the results change.

Am I missing something?


OSR_welfare.mod (4.0 KB)

Dear Dr. Pfeifer,

thanks a lot for your help.

I thought that the algorithm picked by the command opt_algo=9 was a global optimizer. Is that not? and if not, which optimizer would be advisable?

Best Regards

Yes, it’s a global optimizer, but it may still get stuck at a local minimum. For that reason, testing different starting values is exactly what one should do.

Alright, I see, thanks for your help. I will stick with the same algorithm and try different initial values then.

Best Wishes