Particle Filter for second order estimation

Dynare help (legacy posts)
1

Hi,

I am estimating a non linear model using second order approximation on version 4.4.3 and choose option 9 for particle filtering. The problem is that I don’t know at all whether the code is close or far from converging, for example this is one of the lines of output after several runs iteration:

1790 , 23271 : 9.9999955804367e+07 +(2e-02,3e-02) | 6.10e+02 | 2:2.6e-05 19:1.1e-02

How could I possibly find out how far the code is from convergence?

Many thanks,

Z

2

If you have intermediate saving enabled for cmaes, you could use the plotcmaesdat.m from lri.fr/~hansen/cmaes_inmatlab.html to plot the behavior over time and see convergence. The value you posted basically suggests the presence of errors (1e8 is the error code). If all you value are around this, there is something wrong and I might need the mod-file and the data-file.