estimation problem

[scholar_26]

I try to estimate a model. I have tried many trial in estimation command options. But still have some bad posterior results and my MCMC diagnostics graphs do not converge. What can I do? data I used could be wrong?

***************************************************************************************

MH: Multiple chains mode.
MH: Searching for initial values...
MH: Initial values found!

MH: Number of mh files : 46 per block.
MH: Total number of generated files : 92.
MH: Total number of iterations : 250000.
MH: average acceptation rate per chain :
0.1933 0.2691


MCMC Diagnostics: Univariate convergence diagnostic, Brooks and Gelman (1998):
Parameter 1... Done!
Parameter 2... Done!
Parameter 3... Done!
Parameter 4... Done!
Parameter 5... Done!
Parameter 6... Done!
Parameter 7... Done!
Parameter 8... Done!
Parameter 9... Done!
Parameter 10... Done!
Parameter 11... Done!
Parameter 12... Done!
Parameter 13... Done!
Parameter 14... Done!
Parameter 15... Done!
Parameter 16... Done!
Parameter 17... Done!
Parameter 18... Done!
Parameter 19... Done!
Parameter 20... Done!
Parameter 21... Done!

MH: Total number of Mh draws: 250000.
MH: Total number of generated Mh files: 46.
MH: I'll use mh-files 21 to 46.
MH: In mh-file number 21 i'll start at line 3800.
MH: Finally I keep 137500 draws.

MH: I'm computing the posterior mean and covariance... Done!

MH: I'm computing the posterior log marginale density (modified harmonic mean)...
MH: Modified harmonic mean estimator, done!


ESTIMATION RESULTS

Log data density is -1270.416585.

parameters
prior mean post. mean conf. interval prior pstdev

alpha 0.400 0.0653 0.0365 0.0933 beta 0.2000
h 0.700 0.4176 0.3050 0.5171 beta 0.2000
sigma 1.000 -1.0815 -1.3134 -0.8533 norm 0.4000
eta 1.000 1.7733 0.9689 2.5496 gamm 0.4000
phi 1.000 1.6851 1.1920 2.2190 gamm 0.4000
thetah 0.500 0.5590 0.4844 0.6318 beta 0.2500
thetaf 0.500 0.6816 0.6518 0.7095 beta 0.2500
phi1 1.500 1.3123 1.1608 1.4409 gamm 0.2000
phi2 0.250 0.2129 0.1294 0.2892 gamm 0.1000
rhor 0.700 0.3891 0.2966 0.4826 beta 0.2000
rhorst 0.500 0.3985 0.1459 0.6449 beta 0.2000
rhoa 0.500 0.9877 0.9786 0.9972 beta 0.2000
lam1 0.500 0.1539 0.0326 0.2732 beta 0.2000

standard deviation of shocks
prior mean post. mean conf. interval prior pstdev

nu_a 2.000 1.6764 1.2419 2.1326 invg 0.5000
nu_ysts 1.000 1.0541 0.9083 1.2046 invg 0.2000
nu_rsts 0.500 1.0531 0.8811 1.2270 invg 0.2000
nu_pihs 2.000 2.0427 1.7246 2.3700 invg 0.2500
nu_pifs 1.000 2.1352 2.0791 2.1798 invg 0.1000
nu_rs 1.000 1.2694 1.1136 1.4381 invg 0.1000
nu_qs 2.000 2.8317 2.8164 2.8435 invg 0.1000
nu_s 2.000 2.7314 2.6125 2.8435 invg 0.1000
MH: Posterior (dsge) IRFs...
MH: Posterior IRFs, done!
MH: Forecasted variables (mean)
MH: Forecasted variables (mean), done!
MH: Forecasted variables (point)
MH: Forecasted variables (point), done!

[jpfeifer]

Try using a different mode-finder like mode_compute=9. Moreover, consider a longer MC chain and more informative priors.