Dear Statalisters,
I am having trouble using the hotdeck procedure to impute values for y2a
(obs=100, missing=54) from y1 (obs=100, missing=0). The first 25
observations are as follows:
sort y1
list y1 y2a in 1/25
+-----------------------+
| y1 y2a |
|-----------------------|
1. | -1.872524 -1.342797 |
2. | -.950053 . |
3. | -.8294063 . |
4. | -.8209193 2.082552 |
5. | -.6955014 1.37202 |
|-----------------------|
6. | -.542197 . |
7. | -.472739 . |
8. | -.4225743 1.702473 |
9. | -.337009 . |
10. | -.3161629 2.368211 |
|-----------------------|
11. | -.2584664 . |
12. | -.18923 . |
13. | -.1845185 . |
14. | -.1309191 . |
15. | -.1286997 .9998891 |
|-----------------------|
16. | -.12622 3.163509 |
17. | -.1032103 2.781668 |
18. | -.0957911 1.384086 |
19. | -.0614794 1.152725 |
20. | -.0567206 . |
|-----------------------|
21. | -.0264047 .9421599 |
22. | .026992 . |
23. | .0503024 . |
24. | .1294022 . |
25. | .1481918 3.47327 |
+-----------------------+
I use the following code to try to impute for y2a==.
sort y1
hotdeck y2a, by(y1) store
use imp1
list
After the hotdeck command, I get an error message:
"WARNING: 54 strata with NO observed data
46 strata with only 1 observed data
100
WARNING: When the <command> option is not selected
then no analysis is performed on the imputed datasets"
When the list is generated afterwards, no imputation has taken place.
I have no idea how to overcome this. Is it a matter of my data being in long
versus wide format? Any suggestions would be most appreciated.
Reza
*
* 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/
