<>
At the very least, you have to add the -clear- option to the command.
**********
clear*
input cid:mylabel /*
*/score iq nutrit, auto
C1 100 210 200
C1 90 190 180
C2 88 186 150
C2 80 200 210
C3 99 180 199
C3 90 188 211
C1 30 190 201
C1 80 180 190
C2 60 155 176
C2 50 145 145
C3 70 198 134
C3 90 188 122
end
compress
statsby "regress score iq nutrit"/*
*/ _b, by(cid) clear
list, noobs
**********
Apart from that, it does work for me, and leads to
+-----------------------------------------+
cid b_iq b_nutrit b_cons
-----------------------------------------
C1 1.83641 -2.448085 193.3595
C2 .791065 -.2396581 -25.36585
C3 -1.647576 -.0035126 398.4029
+-----------------------------------------+
HTH
Martin
_______________________
----- Original Message -----
From: "Sohini Sahu" <[email protected]>
To: <[email protected]>
Sent: Saturday, June 06, 2009 9:51 PM
Subject: st: Re: Creating a column from point estimates across observations
Thank you Martin.
I was trying out the Statsby command using the following dummy dataset. I
am using version 8 of STATA.
+------------------------------+
| cid score iq nutrit~n |
|------------------------------|
1. | C1 100 210 200 |
2. | C1 90 190 180 |
3. | C2 88 186 150 |
4. | C2 80 200 210 |
5. | C3 99 180 199 |
|------------------------------|
6. | C3 90 188 211 |
7. | C1 30 190 201 |
8. | C1 80 180 190 |
9. | C2 60 155 176 |
10. | C2 50 145 145 |
|------------------------------|
11. | C3 70 198 134 |
12. | C3 90 188 122 |
+------------------------------+
I saved the above in a file.
Then I gave the command
statsby "regress score iq nutrition " _b, by(cid)
I get this
command: regress score iq nutrition
statistics: b_iq = _b[iq]
b_nutrit~n = _b[nutrition]
b_cons = _b[_cons]
by: cid
The data editor has 4 columns now: cid, b_iq, b_nutrit~n and b_cons but
the cells are all empty! Any idea where I am going wrong?
S
*
* 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/
