statalist-digest Saturday, June 23 2007 Volume 04 : Number 2736
** Send unsubscribe or help commands to [email protected] **
The digest contains:
st: error r(7) from outreg
Re: st: error r(7) from outreg
st: poisson regression different data sets-dummy variables v xi
RE: st: version 5.0 to version 9.2
st: RE: poisson regression different data sets-dummy variables v xi
st: Re: smultron text editor
st: Logistic regression, changing order of variables
Re: st: Logistic regression, changing order of variables
st: RE: Logistic regression, changing order of variables
Re: st: Logistic regression, changing order of variables
Re: st: Windows (& Vista) will not search within do files.
Re: st: Windows (& Vista) will not search within do files.
Re: st: Windows (& Vista) will not search within do files.
st: Multinomial endogenous dummy
st: Heckman Model with Endogenous Variable
RE: st: Windows (& Vista) will not search within do files.
st: Adoupdate error
st: Large data sets
st: Re: Large data sets
st: error r(7) from outreg
Wishlist st: Re: Large data sets
----------------------------------------------------------------------
Date: Fri, 22 Jun 2007 03:44:09 -0400
From: "Nirina F" <[email protected]>
Subject: st: error r(7) from outreg
Hello all,
I am not very familiar with outreg yet but following the help and
using it after using ivreg2,
outreg using ivjune22, br addstat(F-stat, e(F), cragg, e(cdchi2),
craggp, e(cdchi2p), craggF, e(cdf), sargan, e(sargan), sarganp,
e(sarga> np))
I get the following message:
e(cdchi2) found where number expected in addstat() option
r(7);
Thanks in advance for your response;
Nirina.
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 10:02:21 +0200
From: "Ben Jann" <[email protected]>
Subject: Re: st: error r(7) from outreg
Not an answer to you question, but maybe try the -esttab- command (it
is part of the -estout- package, see
http://repec.org/bocode/e/estout/):
esttab, brackets scalar("F F-stat" "cdchi2 cragg" "cdchi2p craggp"
"cdf craggF" sargan sarganp)
ben
On 6/22/07, Nirina F <[email protected]> wrote:
> Hello all,
>
> I am not very familiar with outreg yet but following the help and
> using it after using ivreg2,
>
>
> outreg using ivjune22, br addstat(F-stat, e(F), cragg, e(cdchi2),
> craggp, e(cdchi2p), craggF, e(cdf), sargan, e(sargan), sarganp,
> e(sarga> np))
>
> I get the following message:
>
> e(cdchi2) found where number expected in addstat() option
> r(7);
> Thanks in advance for your response;
>
> Nirina.
> *
> * 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/
>
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 09:33:46 +0000 (GMT)
From: ANN FITZMAURICE <[email protected]>
Subject: st: poisson regression different data sets-dummy variables v xi
hi
i know that i can use the following
Use data1 , clear
Poisson y x
Predict file1_m, ir
Use data 2, clear
Predict file2_m , ir
My question relates to the use of the command xi, in the regression the command is
xi:poisson var1 i.var2 i.var3 var4 etc.
but when i try to apply the results from file 1 to file 2 , i get an error message, for this to work prior to any analysis would i have to create dummy variables in b oth of the data sets . I plan on running a series of regressions using different combnation of the independent vars and am therfore trying to automated the procedure as much as possible.
I could of course run all of the regressions in file one and output the seperate predicted varaibles to the second file, but thought using the above would be more efficient
this analyses will be repeated on several different country data sets, therefore looking for an efficient method
I trust that this explanation is ok
ann
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 10:54:17 +0100
From: "Nick Cox" <[email protected]>
Subject: RE: st: version 5.0 to version 9.2
True enough, but if you're cycling over a numlist 1/`r'
you might as well translate this directly to
inspect `2'
local r=`r(N_unique)' - 1
forval v = 1/`r' {
matrix C[`v',1]=c`v'
}
and then this is a smidgen cleaner:
inspect `2'
local r = r(N_unique) - 1
forval v = 1/`r' {
matrix C[`v',1] = c`v'
}
and this is a smidgen shorter still,
although perhaps more difficult to read:
inspect `2'
forval v = 1/`= r(N_unique) - 1' {
matrix C[`v',1] = c`v'
}
But the code is all followed by a -matrix list C-.
And those -c1-, -c2-, etc. are presumably scalars,
that must exist already. So I don't understand this
really, I am just translating code.
Nick
[email protected]
P.S. I guess the -nostop- would be translated
by a -capture- somewhere.
Abdel Rahmen El Lahga
>
> inspect `2'
> local r=`r(N_unique)' - 1 // maxt is ignored
> foreach v of numlist 1/`r' {
> matrix C[`v',1]=c`v'
> }
> ltype(numeric) is to tell Stata that the cooresponding list is numeric
> not varlist
> nostop tell Stata to not stop the loop even if there are an error but
> i don't know to deal with this option
> HTH
> AbdelRahmen El Lahga
> 2007/6/22, Raphael Fraser <[email protected]>:
> > An additional question:
> >
> > inspect `2'
> > local maxt=`r(N_unique)'
> > local r=`maxt'-1
> > for 1-`r', ltype(numeric) nostop: matrix C[@,1]=c@ /* <--
> Convert to
> > Stata 9 code */
> > matrix list C
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 13:50:16 +0200
From: "Maarten Buis" <[email protected]>
Subject: st: RE: poisson regression different data sets-dummy variables v xi
- -- ANN FITZMAURICE wrote
> i know that i can use the following
>
> Use data1 , clear
> Poisson y x
> Predict file1_m, ir
> Use data 2, clear
> Predict file2_m , ir
>
> My question relates to the use of the command
> xi, in the regression the command is
> xi:poisson var1 i.var2 i.var3 var4 etc.
>
> but when i try to apply the results from file 1
> to file 2 , i get an error message
The reason you get an error is that -xi- creates
new variables, and these variables are used in
your estimation command, but they are not created
in you second file. So what you need to do is to
make the same variables in data2, like in the
example below.
*------------- begin example -------------
sysuse auto, clear
recode rep78 1/2=3
/* create two files */
gen select = uniform() < .5
tempfile data1 data2
preserve
keep if select == 0
save `data1'
restore
keep if select == 1
save `data2'
/* Estimation */
use `data1', clear
xi i.rep78 /*create dummies*/
reg mpg length _I*
/* Prediction */
use `data2', clear
xi i.rep78 /*create dummies*/
predict xb, xb
*-------------- end example --------------
(For more on how to use examples I sent to the Statalist, see
http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
Hope this helps,
Maarten
- -----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
- -----------------------------------------
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 08:15:46 -0400
From: Kit Baum <[email protected]>
Subject: st: Re: smultron text editor
That Smultron preference is meant to point to an executable program,
not to a directory. /bin/sh is a program (one flavor of the Unix
shell). /applications/stata/stata-se is, likewise, a Unix program
(the console version of Stata/SE 9). I believe you need the <
redirection on the line as well to get Smultron to feed what you have
highlighted into Stata.
Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html
On Jun 22, 2007, at 2:33 AM, Michael wrote:
> I've altered Smultron's Advanced Preference for Run Text to my stata
> path, which is:
> /applications/stata
> However, when I Run Text, the results window comes up blank. I've
> contacted the Smultron developer, and he is not familiar with Stata.
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 15:33:45 +0200
From: "Todd Swarthout" <[email protected]>
Subject: st: Logistic regression, changing order of variables
Hello everyone.
I am using logistic regression in the analysis of a database coming from
a study in Zimbabwe. A sample command is 'xi:logistic hiv i.age3 sexc
age'
The variable age3 is the age in 3 strata (2-4yrs, 5-11yrs, 12-17yrs). As
I currently have the coding, 0=2-4yrs, 1=5-11yrs, and 2=12-17yrs. Thus
the strata 2-4yrs is the point of comparison. In this case, the two
older strata both showed a reduced risk of infection.
However, I prefer to report the data in the context of which strata has
increased risk of infection. Is there a command I can use (or add to the
existing command 'xi:logistic hiv i.age3 sexc age') that would allow me
to change the order of strata during logistic analysis. That is to say,
if I want to use the 12-17 yr strata as the point of
comparison...without having to create a new variable in which I change
the coding.
Thank you
Todd
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 14:46:53 +0100
From: "Neil Shephard" <[email protected]>
Subject: Re: st: Logistic regression, changing order of variables
On 6/22/07, Todd Swarthout <[email protected]> wrote:
> Hello everyone.
>
> I am using logistic regression in the analysis of a database coming from
> a study in Zimbabwe. A sample command is 'xi:logistic hiv i.age3 sexc
> age'
>
> The variable age3 is the age in 3 strata (2-4yrs, 5-11yrs, 12-17yrs). As
> I currently have the coding, 0=2-4yrs, 1=5-11yrs, and 2=12-17yrs. Thus
> the strata 2-4yrs is the point of comparison. In this case, the two
> older strata both showed a reduced risk of infection.
>
> However, I prefer to report the data in the context of which strata has
> increased risk of infection. Is there a command I can use (or add to the
> existing command 'xi:logistic hiv i.age3 sexc age') that would allow me
> to change the order of strata during logistic analysis. That is to say,
> if I want to use the 12-17 yr strata as the point of
> comparison...without having to create a new variable in which I change
> the coding.
>
See -man xi- pages under the heading "Summary of controlling the
omitted dummy" to alter which dummy variable is omitted and used as
reference.
Neil
- --
"In mathematics you don't understand things. You just get used to
them." - Johann von Neumann
Email - [email protected] / [email protected]
Website - http://slack.ser.man.ac.uk/
Photos - http://www.flickr.com/photos/slackline/
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 15:47:22 +0200
From: "Maarten Buis" <[email protected]>
Subject: st: RE: Logistic regression, changing order of variables
- --- Todd Swarthout wrote:
> Is there a command I can use (or add to the existing command
> -xi:logistic hiv i.age3 sexc age-) that would allow me to change
> the order of strata during logistic analysis. That is to say,
> if I want to use the 12-17 yr strata as the point of
> comparison...without having to create a new variable in which I
> change the coding.
It can be done with -xi-, see -help xi- under the section
"Summary of controlling the omitted dummy". So you would type:
char age3[omit] 2
xi:logistic hiv i.age3 sexc age
I like to create my own dummies. That way I have more control
over the reference category, any other contrast I may like, and
the variable names. So I would prefer to do:
gen age2_4 = age3 == 0 if age3 < .
gen age5_11 = age3 == 1 if age3 < .
logistic hiv age2_4 age5_11 sexc age
Hope this helps,
Maarten
- -----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
- -----------------------------------------
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 15:59:17 +0200
From: Christian Ganser <[email protected]>
Subject: Re: st: Logistic regression, changing order of variables
char age3 [omit] 2
Todd Swarthout wrote:
> Hello everyone.
>
> I am using logistic regression in the analysis of a database coming from
> a study in Zimbabwe. A sample command is 'xi:logistic hiv i.age3 sexc
> age'
>
> The variable age3 is the age in 3 strata (2-4yrs, 5-11yrs, 12-17yrs). As
> I currently have the coding, 0=2-4yrs, 1=5-11yrs, and 2=12-17yrs. Thus
> the strata 2-4yrs is the point of comparison. In this case, the two
> older strata both showed a reduced risk of infection.
>
> However, I prefer to report the data in the context of which strata has
> increased risk of infection. Is there a command I can use (or add to the
> existing command 'xi:logistic hiv i.age3 sexc age') that would allow me
> to change the order of strata during logistic analysis. That is to say,
> if I want to use the 12-17 yr strata as the point of
> comparison...without having to create a new variable in which I change
> the coding.
>
> Thank you
>
> Todd
>
> *
> * 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/
>
>
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 16:00:06 +0100
From: Paul Seed <[email protected]>
Subject: Re: st: Windows (& Vista) will not search within do files.
Thanks to all who responded to this.
I have found a workable solution (for me).
Of the various solutions suggested,
the best (for me) seems to be Grep.
(downloaded from http://www.wingrep.com/)
As far as I can see, it is not possible to configure
the search type within Windows XP as it is with Vista.
The free versions of Ultra-Edit & WinEdt are a time-limited.
(unlike Grep, which merely suggests you register.)
-find- from a DOS prompt gives too much text
(including files without the text searched for)
So thanks in particular to Al Feiveson's son.
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 16:24:11 +0100
From: "Neil Shephard" <[email protected]>
Subject: Re: st: Windows (& Vista) will not search within do files.
On 6/22/07, Paul Seed <[email protected]> wrote:
> Thanks to all who responded to this.
> I have found a workable solution (for me).
>
> Of the various solutions suggested,
> the best (for me) seems to be Grep.
> (downloaded from http://www.wingrep.com/)
>
> As far as I can see, it is not possible to configure
> the search type within Windows XP as it is with Vista.
> The free versions of Ultra-Edit & WinEdt are a time-limited.
> (unlike Grep, which merely suggests you register.)
> -find- from a DOS prompt gives too much text
> (including files without the text searched for)
If you wanted to avoid WinGrep's suggestion to register you could go
the whole hog, and use the original grep by installing Cygwin and
searching your files from within a UNIX-like shell.
(If I remember at the start of the thread you mentioned that you
didn't want to install a different OS, this provides the functionality
of an alternative OS under M$-Windows).
Neil
- --
"In mathematics you don't understand things. You just get used to
them." - Johann von Neumann
Email - [email protected] / [email protected]
Website - http://slack.ser.man.ac.uk/
Photos - http://www.flickr.com/photos/slackline/
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 12:58:46 -0400
From: David Kantor <[email protected]>
Subject: Re: st: Windows (& Vista) will not search within do files.
At 04:55 PM 6/21/2007, Rodrigo wrote:
>I like your solution. I should say rather than complex, over-protected
>over-"visualized". Maybe someday we will return to the basics.
>
>DOS lives!! :-)
>Rodrigo.
>[...]
- --in response to my message:
>Several people have given useful suggestions. I will tell how I
>handle this. I do it the old-fashioned way -- in the command prompt
>using the -find- command. (And I never had the problem Paul
>mentioned; I wasn't even aware of it.) I may be out-of-date, but to
>me this is an example of how modern user-interfaces have become so
>complex that they cause more problems than they solve.
Thank you for that validation. And I agree with your assessment of
"over-protected" and "over-visualized".
I'm not really a champion of DOS in particular. (And what we are
dealing with is really the command environment of Windows, which is
not exactly the same as DOS.) I also find DOS (or the Windows
command environment) to be a bit disappointing, after having worked
in some others for many years. What I do prefer is the general
concept of a command environment -- old-fashioned, but very powerful
if you know how to use it. Evidently, many others agree, as some of
the other responses have been about programs that provide various
alternative command environments under Windows.
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 13:23:26 -0400
From: "Hinh Khieu" <[email protected]>
Subject: st: Multinomial endogenous dummy
Dear All:
I have a multinomial dummy variable that has 7 categories - coded 0 to 6.
This dummy is an endogenous explanatory variable in a model like the
following:
Y = a0 + a1*MultinomialDummy + a2*ExogenousVariables + errorterms
My question is how I should account for this dummy using STATA? All I guess
is that I should use mlogit to run the first-stage regression. I have no
clue what to get from the mologit to plug in the second-stage. I don't think
it's just a simple predicted value.
Since I am interested in the impact of each of the category of the
MultinomialDummy on my Y, some suggestions were made that I run the
following model instead:
Y = a0 + a1*Dummy1 + a2Dummy1*S1 + ... + a12*Dummy6 + a13*Dummy6*S6 +
a15*ExogenousVariables + errorterms
where Dummy1 is (1,0) when one of the category of the MultinomialDummy
variable above occurs, and so on for Dummy2 to Dummy6.
What S1 to S6 that are interacted with the dummies are a big puzzle to me.
Does anyone know what they are and how to get them by STATA or mlogit?
I'd greatly appreciate any comments on the econometric method and STATA.
Thanks,
Hinh
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 11:13:57 -0700 (PDT)
From: "Ana R. Rios" <[email protected]>
Subject: st: Heckman Model with Endogenous Variable
Dear Stata users,
I would appreciate any advise/help regarding the use
of a Heckman model with an endogenous variable.
Thanks,
Ana Rios
____________________________________________________________________________________
Take the Internet to Go: Yahoo!Go puts the Internet in your pocket: mail, news, photos & more.
http://mobile.yahoo.com/go?refer=1GNXIC
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 14:20:30 -0500
From: "Grealy, Patrick J" <[email protected]>
Subject: RE: st: Windows (& Vista) will not search within do files.
In addition to the DOS -find- command there is -findstr- which I use
quite frequently on my Windows XP machine. Even for file types which the
Windows Search includes, I prefer the output of this command when used
on text files. The -findstr- command has many more options, including -
/s -, which searches through all sub-directories. There is also
provision for handling regular expressions and pattern matching but I
have never used these and cannot say how well they perform.
Pat G.
- -----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of David Kantor
Sent: Friday, June 22, 2007 11:59 AM
To: [email protected]
Subject: Re: st: Windows (& Vista) will not search within do files.
At 04:55 PM 6/21/2007, Rodrigo wrote:
>I like your solution. I should say rather than complex, over-protected
>over-"visualized". Maybe someday we will return to the basics.
>
>DOS lives!! :-)
>Rodrigo.
>[...]
- --in response to my message:
>Several people have given useful suggestions. I will tell how I
>handle this. I do it the old-fashioned way -- in the command prompt
>using the -find- command. (And I never had the problem Paul
>mentioned; I wasn't even aware of it.) I may be out-of-date, but to
>me this is an example of how modern user-interfaces have become so
>complex that they cause more problems than they solve.
Thank you for that validation. And I agree with your assessment of
"over-protected" and "over-visualized".
I'm not really a champion of DOS in particular. (And what we are
dealing with is really the command environment of Windows, which is
not exactly the same as DOS.) I also find DOS (or the Windows
command environment) to be a bit disappointing, after having worked
in some others for many years. What I do prefer is the general
concept of a command environment -- old-fashioned, but very powerful
if you know how to use it. Evidently, many others agree, as some of
the other responses have been about programs that provide various
alternative command environments under Windows.
*
* 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/
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 20:26:22 +0100
From: "Nikolaos Kanellopoulos" <[email protected]>
Subject: st: Adoupdate error
I run adoupdate, all update a few day ago and now when I run adoupdate,
update I get the following:
. adoupdate, update
(note: adoupdate updates user-written files; type -update- to check for
updates to official Stata)
directory does not have user-installed files
r(601);
Can anybody enlight me?
Nikos
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 16:56:43 -0500
From: "Carmen Ponce" <[email protected]>
Subject: st: Large data sets
Hi,
I need to work with large datasets (>700M, <2g), but Stata 9.2 does not
allow me to set memory above 500M (my computer should allow for up to 2g
memory setting).
I checked on Stata�s FAQs website
(http://www.stata.com/support/faqs/win/winmemory.html) and found out about
the hotfix patch 894472 that helps overcome this problem. I tried to execute
the hotfix patch from this website but got this message "Setup cannot update
Windows XP files because the language installed on your system is different
from the update language".
I tried to find a version in spanish (my system is set in spanish) but have
not succeded..
Does anyone know about a spanish version of this patch?
Thank you in advance,
Carmen
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 18:13:48 -0400
From: "Michael Blasnik" <[email protected]>
Subject: st: Re: Large data sets
....
I don't know about a Spanish language version of the patch, but even without the
patch you should be able to allocate about 900m to Stata under XP. How much
physical memory do you have? Do you have lots of other programs running at the
same time? You may want to reduce the number of programs that start up
automatically and launch Stata with nothing else running to see if you can get
to something close to 900m.
There's still a good chance that you will actually need more than 900m if you
really need to use a 700m+ dataset. But you may be able to make the dataset
smaller: Can you drop some variables? Could you compress some variables using
value labels or smaller storage types or even using abbreviations in text
strings? Can you drop some observations for some analyses? All of these
options are worth exploring. Otherwise, you could switch to an OS that can
allocate more memory than XP -- Mac, Linux, even Windows Vista can allocate
more.
Michael Blasnik
- ----- Original Message -----
From: "Carmen Ponce" <[email protected]>
To: <[email protected]>
Sent: Friday, June 22, 2007 5:56 PM
Subject: st: Large data sets
> Hi,
> I need to work with large datasets (>700M, <2g), but Stata 9.2 does not
> allow me to set memory above 500M (my computer should allow for up to 2g
> memory setting).
> I checked on Stata�s FAQs website
> (http://www.stata.com/support/faqs/win/winmemory.html) and found out about
> the hotfix patch 894472 that helps overcome this problem. I tried to execute
> the hotfix patch from this website but got this message "Setup cannot update
> Windows XP files because the language installed on your system is different
> from the update language".
> I tried to find a version in spanish (my system is set in spanish) but have
> not succeded..
> Does anyone know about a spanish version of this patch?
>
> Thank you in advance,
> Carmen
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 17:58:34 -0700
From: "roy wada" <[email protected]>
Subject: st: error r(7) from outreg
Nirina wants outreg to work with ivreg2. Her codes does not work
because ivreg2 does not always return a number in its scalars. In
other words, e(cdchi2) is probably missing from her regression result.
The following codes will work.
sysuse auto,clear
ivreg2 mpg price (headroom=rep78)
outreg using ivjune22, br addstat(F-stat, e(F), sargan, e(sargan))
The following code will not work because some of the scalars are
missing from ivreg2 scalars.
outreg using ivjune22, br addstat(F-stat, e(F), cragg, e(cdchi2), craggp,
e(cdchi2p), craggF, e(cdf), sargan, e(sargan), sarganp, e(sarganp))
To see what is contained or missing from ivreg2 scalars, type these:
ssc install outreg2, replace
outreg2 using ivjune22, br e(all) seeout
Her new codes would look like this:
outreg2 using ivjune22, br addstat(F-stat, e(F), cragg, e(cdchi2), craggp,
e(cdchi2p), craggF, e(cdf), sargan, e(sargan), sarganp, e(sarganp))
seeout
Roy
>I am not very familiar with outreg yet but following the help and
>using it after using ivreg2,
>
>
>outreg using ivjune22, br addstat(F-stat, e(F), cragg, e(cdchi2),
>craggp, e(cdchi2p), craggF, e(cdf), sargan, e(sargan), sarganp,
>e(sarga> np))
>
>I get the following message:
>
>e(cdchi2) found where number expected in addstat() option
>r(7);
>Thanks in advance for your response;
>
>Nirina.
_________________________________________________________________
Need a break? Find your escape route with Live Search Maps.
http://maps.live.com/default.aspx?ss=Restaurants~Hotels~Amusement%20Park&cp=33.832922~-117.915659&style=r&lvl=13&tilt=-90&dir=0&alt=-1000&scene=1118863&encType=1&FORM=MGAC01
*
* 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/
------------------------------
Date: Fri, 22 Jun 2007 22:17:59 -0700 (PDT)
From: SamL <[email protected]>
Subject: Wishlist st: Re: Large data sets
This brings up a general problem that I wonder if stata can fix, or has
fixed. I routinely use large datasets--by large I mean 4-10 gb. I use a
unix system managed by a computer center. Sometimes my data becomes so
large that there is no machine large enough to invoke stata and hold all
the data in memory.
I should also say that my models routinely take weeks or even months to
run, so I am resigned to waiting a long time for results. Speed is not my
interest.
So, I was wondering whether stata can, or has, made it possible to tell
stata to use a disk as virtual memory. I know this will slow down run
times substantially, and my 4 week job could easily become a 6 month job.
But, at present, my 4 week job cannot be run without more resources, and
that means waiting a year to get more money to buy more memory. So, even
at 6 months, that would be more than twice as fast as now.
I know this is not the common desire--I chuckle when someone complains
that it took 57 seconds to run something rather than the 35 seconds they
had expected, and wonder whether stata is efficient enough. Yes,
efficiency there would help me, too. But what would help far more is the
ability to just opt out of holding all the data in memory when I need to
do so.
FYI--Yes, I've done all the data reduction one might suggest (e.g., using
frequency weights for patterns in the data, and so forth) and no, for my
problem sampling will not be helpful because the models I am estimating
need all the data to address issues of sparseness in some parts of the
data, and sampling will prevent that gain in information.
So, how 'bout it stata--have you made it possible for one to use a disk as
virtual memory with a few commands and, if not, will you please do so?
Respectfully yours,
Sam
On Fri, 22 Jun 2007, Michael Blasnik wrote:
> ....
> I don't know about a Spanish language version of the patch, but even without the
> patch you should be able to allocate about 900m to Stata under XP. How much
> physical memory do you have? Do you have lots of other programs running at the
> same time? You may want to reduce the number of programs that start up
> automatically and launch Stata with nothing else running to see if you can get
> to something close to 900m.
>
> There's still a good chance that you will actually need more than 900m if you
> really need to use a 700m+ dataset. But you may be able to make the dataset
> smaller: Can you drop some variables? Could you compress some variables using
> value labels or smaller storage types or even using abbreviations in text
> strings? Can you drop some observations for some analyses? All of these
> options are worth exploring. Otherwise, you could switch to an OS that can
> allocate more memory than XP -- Mac, Linux, even Windows Vista can allocate
> more.
>
> Michael Blasnik
>
> ----- Original Message -----
> From: "Carmen Ponce" <[email protected]>
> To: <[email protected]>
> Sent: Friday, June 22, 2007 5:56 PM
> Subject: st: Large data sets
>
>
> > Hi,
> > I need to work with large datasets (>700M, <2g), but Stata 9.2 does not
> > allow me to set memory above 500M (my computer should allow for up to 2g
> > memory setting).
> > I checked on Stata�s FAQs website
> > (http://www.stata.com/support/faqs/win/winmemory.html) and found out about
> > the hotfix patch 894472 that helps overcome this problem. I tried to execute
> > the hotfix patch from this website but got this message "Setup cannot update
> > Windows XP files because the language installed on your system is different
> > from the update language".
> > I tried to find a version in spanish (my system is set in spanish) but have
> > not succeded..
> > Does anyone know about a spanish version of this patch?
> >
> > Thank you in advance,
> > Carmen
>
> *
> * 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/
>
*
* 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/
------------------------------
End of statalist-digest V4 #2736
********************************
*
* 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/
