[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: A Problem on generating a variable using --appended-dataset

From   "Nick Cox" <>
To   <>
Subject   st: RE: A Problem on generating a variable using --appended-dataset
Date   Mon, 19 Nov 2007 10:17:26 -0000

You can't -generate- a variable that already exists. 
That is basic: read about -generate- and -replace-. 

You may want to -generate- outside the loop
and -replace ... in `i'- within the loop.

gen X = . 
forval i = 44/93 { 
	replace ... in `i' 

Asgar Khademvatani

Basically, I am estimating a cost model using time series dataset over
1958-2000. Then, I have appended a new observation over 2001-2050 to my
master dataset in the Stata memory, using--append-- command. I want to
generate some new variables defined based on functions upon estimated
coefficient and the other variables, I want generate these variables
only over 2001-2050 during the same period of new using
observation.Actually, I already tried --gen-- command but it only
generate the variables over master dataset period over 1958-2000 NOT
over 2001-2050. In doing so, I am going to write a --forvalues-- loop to
do so. But the variables only are generated at one single
observation,say only 2001, and it is re-extended over other years with
the same values . For example, my written loop and the stata results are
given as:
A Sample Loop:

 forvalues i = 44/93 {
     gen X = _b[le_sg] + _b[lple_sg]*(plpl_sg[`i']) +
_b[lpme_sg]*(plpm_sg[`i']) + _b[lek_neg_sg]*(plk_sg[`i']) +
     list  X

The Stata Output:

 do "C:\DOCUME~1\ASGARK~1\LOCALS~1\Temp\STD0i000000.tmp"

.  1.  forvalues i = 44/93 {
  2.      gen X =_b[le_sg] + _b[lple_sg]*(plpl_sg[`i']) + 
_b[lpme_sg]*(plpm_sg[`i']) + _b[lek_neg_sg]*(plk  > _sg[`i']) +
  3.      list  X
  4. }        

     |       X
  1. | -7.730469 |
 94. | -7.730469 |
X already defined

end of do-file

As you see in the above, X variable is generated only in 2001 and it is
repeated over next years with the same value. I am not sure what is
wrong with my loop.  May somebody please help me out on this matter and
tell me what is wrong with that? Otherwise, I am just wondering that is
there another way other than writing a loop for generating the variables
over 2001-2050?

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index