Timeline for Reshaping/transposing data in Stata
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 5 at 15:28 | comment | added | Nick Cox | I've been a member of Statalist since 1994 and never seen a random question. Perhaps introduced later. | |
| Jul 5 at 15:15 | vote | accept | Emma | ||
| Jul 5 at 15:38 | |||||
| Jul 5 at 15:04 | history | edited | Nick Cox | CC BY-SA 4.0 |
added 1388 characters in body
|
| Jul 5 at 15:04 | comment | added | Nick Cox | I don't follow that. If you can talk to SO, you can talk to Statalist. But any way, please see my edit. | |
| Jul 5 at 15:01 | comment | added | Emma | I'm unfortunately unable to register for statalist as my version of stata is based on a server without internet access, and so cannot get the authkey to register or contact them! | |
| Jul 5 at 14:46 | comment | added | Emma | Sure I've edited my question to reflect this. But I think you might be right, I have also been thinking I may need to look into a different data layout! | |
| Jul 5 at 14:44 | comment | added | Nick Cox | That said, I fear that you're asking for a data layout that won't work for your data. I would post a version of this on Statalist where you're much more likely to get attention from medical statisticians. | |
| Jul 5 at 14:32 | comment | added | Nick Cox | You need to revise your question so that you show a dataset that is complicated enough to show the problem and compact enough to show here. | |
| Jul 5 at 14:00 | comment | added | Emma |
Hi Nick, Thanks so much for your speedy help. I do still have a problem however, which I think is due to having duplicated IDs in the controlb1-b4 columns, and differing indexdates for these. (Some controls were matched to multiple cases, and the indexdate is decided based on the case.) Is there any way around this? Thanks. Error: variable id does not uniquely identify the observations Your data are currently wide. You are performing a reshape long. You specified i (index_date) and j(which). In the current wide form, variable index_date should uniquely identify the observations.
|
|
| Jun 11 at 6:27 | history | edited | Nick Cox | CC BY-SA 4.0 |
added 342 characters in body
|
| Jun 10 at 20:54 | history | edited | Nick Cox | CC BY-SA 4.0 |
edited body
|
| Jun 10 at 19:44 | history | answered | Nick Cox | CC BY-SA 4.0 |