Acceptance ratios, ratio and rate

[gin]

Dear all,

When working in a console mode, I'm seeing Dynare (4.4.0) reporting Acceptance ratios, ratio and rate in a single session. Is there a fundamental difference between those three terms, or otherwise we might use just one of them?

gin

[jpfeifer]

They are the same. Could you provide the output in question so that we can harmonize it?

[gin]

Those words do not appear in the log-file (even with the `debug' option), but just in the Matlab's console as such (before I pressed `ctrl+C'):

Code: Select all

Loading 71 observations from ctwData8.m

Initial value of the log posterior (or likelihood): -6484.7307
Acceptance ratios: 0.280833 [0.316000]   3% done
Acceptance ratio: 0.377300 100% done

==========================================================
   Change in the covariance matrix = 9.9999.
   Mode improvement = 3229.2936
   New value of jscale = 0.0039855
==========================================================
Acceptance ratios: 0.490400 [0.310000]   3% done
Acceptance ratio: 0.296550 100% done

==========================================================
   Change in the covariance matrix = 1.3393.
   Mode improvement = 39.6372
   New value of jscale = 0.074675
==========================================================
Acceptance ratios: 0.362667 [0.328000]   2% done
Acceptance ratio: 0.337350 100% done
Acceptance rates: 0.372300 [0.343000]   5% done
1.000000 Jumps / MaxStepSize 0.051014  35% done

The estimation command is the following:

Code: Select all

options_.console_mode=1; %(default: 0)
estimation(order=1,datafile=ctwData8, mh_replic=900000, mh_nblocks=1, mh_drop=.3,  mh_jscale=0.2,
 mode_compute=6, filter_step_ahead = [1 4 8 12], forecast=20,smoother, filtered_vars, diffuse_filter,
nodisplay, moments_varendo, conditional_variance_decomposition=[1 4 8 20],
graph_format=(eps,pdf));

gin

[jpfeifer]

Thanks for pointing this out. I will harmonize it.