Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: AW: Keeping variable After restore


From   Eric Booth <[email protected]>
To   [email protected]
Subject   Re: st: AW: Keeping variable After restore
Date   Mon, 9 Nov 2009 08:41:44 -0600

Or if you insist on using -preserve-/-restore-, you can create a new dataset out of the preserved data & -merge- it (though I think Martin's solution using conditionals is better, especially if your dataset is really large--saving a new dataset and merging it back into the master would be slower):

*****
clear
sysuse auto
expand 2
gen year = 2000 in 1/50
replace year = 2001 in 50/l
sort make year
**
preserve
keep if year == 2000
factor mpg price rep78
predict var1
keep make year var1
sort make year
save formerge.dta, replace
restore

merge make year using formerge.dta
browse make year var1
*****

Eric

__
Eric A. Booth
Public Policy Research Institute
Texas A&M University
[email protected]
Office: +979.845.6754


On Nov 9, 2009, at 8:28 AM, Martin Weiss wrote:


<>

Instead of -keep-ing the year 2000:

***
webuse bg2, clear
gen year=2000*(_n<200)+ /*
*/ 2001*(_n>=200)
qui factor bg2cost1-bg2cost6  /*
*/ if year==2000
predict var1
***

and off you go...



HTH
Martin


-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von C.T. Shehzad
Gesendet: Montag, 9. November 2009 15:22
An: [email protected]
Betreff: st: Keeping variable After restore

Dear Stata-listers:

I want to generate a variable during (preserve - restore) and want to
use that after restoring the data. When I will restore data the variable
will be dropped automatically. How can I keep the variable?

The data is time series and a simple version of my program is as follows:

preserve
keep if year == 2000
factor x1 x2 x3
predict var1
restore

gsort -year
replace var1= var1[_n-1] if var1 >= .
xtset country year
xtreg dep var1 controls

The problem in this program is that after restoring the data I will not
have var1 anymore. What can be a possible solution?

Thanks and best regards,

Tanveer



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index