Dear Statalisters,
I am experiencing some unexpected behaviour from my fully updated Stata 9
running on Windows XP. I wonder if anyone can confirm the following. It's
about this simple program:
clear
sysuse auto
set tracedepth 1
set tracenumber on
set trace on
logistic foreign headroom
If I -do- the program from the do-file editor, I get this result:
----------------------------------------------------------------------------
------------------------------------------------------ begin logistic ---
02 - version 6.0, missing
02 - local version : di "version " string(_caller()) ", missing :"
02 - if _by() {
02 local BY `"by `_byvars'`_byrc0':"'
02 }
02 - `version' `BY' _vce_parserun logistic : `0'
= version 9, missing : _vce_parserun logistic : foreign headroom
02 - if "`s(exit)'" != "" {
= if "" != "" {
02 exit
02 }
02 - if replay() {
02 if `"`e(cmd)'"'~=`"logistic"' { error 301 }
02 if _by() { error 190 }
02 Display `0'
02 exit
02 }
02 - quietly syntax [anything] [if] [in] [fw iw pw aw] [, * ]
02 - `version' `BY' Estimate `0'
= version 9, missing : Estimate foreign headroom
Logistic regression Number of obs =
74
LR chi2(1) =
6.72
Prob > chi2 =
0.0095
Log likelihood = -41.671379 Pseudo R2 =
0.0747
----------------------------------------------------------------------------
--
foreign | Odds Ratio Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
headroom | .4304066 .1490467 -2.43 0.015 .2183303
.8484845
----------------------------------------------------------------------------
--
----------------------------------------------------------------------------
-------------------------------------------------------- end logistic ---
If, on the other hand, I copy and paste the lines into the command window, I
get this (which is what I expected):
----------------------------------------------------------------------------
-------------------------------------------------------- begin logistic ---
01 - version 6.0, missing
01 - local version : di "version " string(_caller()) ", missing :"
01 - if _by() {
01 local BY `"by `_byvars'`_byrc0':"'
01 }
01 - `version' `BY' _vce_parserun logistic : `0'
= version 9, missing : _vce_parserun logistic : foreign headroom
01 - if "`s(exit)'" != "" {
= if "" != "" {
01 exit
01 }
01 - if replay() {
01 if `"`e(cmd)'"'~=`"logistic"' { error 301 }
01 if _by() { error 190 }
01 Display `0'
01 exit
01 }
01 - quietly syntax [anything] [if] [in] [fw iw pw aw] [, * ]
01 - `version' `BY' Estimate `0'
= version 9, missing : Estimate foreign headroom
Logistic regression Number of obs =
74
LR chi2(1) =
6.72
Prob > chi2 =
0.0095
Log likelihood = -41.671379 Pseudo R2 =
0.0747
----------------------------------------------------------------------------
--
foreign | Odds Ratio Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
headroom | .4304066 .1490467 -2.43 0.015 .2183303
.8484845
----------------------------------------------------------------------------
--
----------------------------------------------------------------------------
---------------------------------------------------------- end logistic ---
These two outputs are similar except from the numbering and tabbing, I'd
say.
If, in the do-file editor, I change -set tracedepth 1- to -set tracedepth
3-, the tracenumbers will range from 2 to 4. If I do the same from the
command window, I get the expected range of 1 to 3.
I'm guessing that there is something odd going on with the tracenumbering.
Best regards,
Peter.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
