Dear Friends,
I have a question about SAS PROC MCMC. Below is an excerpt of an output
I got.
Monte Carlo Standard Errors
Standard
Parameter MCSE Deviation MCSE/SD
th1 0.000141 0.0117 0.0120
th2 0.00424 0.3784 0.0112
th3 0.00389 0.3356 0.0116
Heidelberger-Welch Diagnostics
Stationarity Test Half-width Test
Cramer-von- Test Iterations
Half- Relative Test
Parameter Mises Stat p Outcome Discarded width Mean
Half-width Outcome
th1 0.2305 0.2154 Passed 0
0.000254 1.8102 0.000140 Passed
th2 0.4201 0.0640 Passed 0 0.00747 -7.9662
-0.00094 Passed
th3 0.2962 0.1385 Passed 0 0.00752 -1.0051
-0.00748 Passed
Now, if I understand the theory and the documentation correctly, the
Half-width in the Heidelberger-Welch panel should equal 1.96 times the MCSE.
But in this output it does not come out this way (the results are close,
but different enough not to be due to rounding error).
If anyone on the list can explain to me what's going on, I'd be grateful.
With thanks and regards,
David
--
------------------------------------------------------------------------
David M. Zucker
Department of Statistics, Hebrew University
Mt. Scopus, Jerusalem ISRAEL
Phone: +972-2-5881291, FAX: +972-2-588-3549
E-mail: [log in to unmask]
<mailto:[log in to unmask]>Website:
http://pluto.huji.ac.il/~mszucker/ <http://pluto.huji.ac.il/%7Emszucker/>
HU Stat Dept Web Page: http://www.stat.huji.ac.il/
HU Stat Dept Facebook Page:
http://www.facebook.com/home.php?#!/group.php?gid=100748143315636
<http://www.facebook.com/home.php?#%21/group.php?gid=100748143315636>
------------------------------------------------------------------------
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|