Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

re: st: Merge/Joinby Question

From   "Airey, David C" <>
To   "" <>
Subject   re: st: Merge/Joinby Question
Date   Wed, 6 Jul 2011 14:25:19 -0500


I just spent a few days manipulating data with merges and appends etc. and 
I can't follow your question in enough detail to think of a solution. I may
just not be getting it. Sometimes showing a mock data set or subsets of your 
data help to spark a clear solution.

> Hi -
> I have data from a propensity score match that has duplicate ids in it (data set A),
> because it was with replacement. However, now I am interested in merging
> data set A with an unbalanced panel data set (data set B).  So, neither data set has
> a unique identifier.   
> Is joinby the correct command to ensure that each person (duplicated ids are 
> assumed to be "different" people in my post-match world) has the correct number 
> of observations?
> One way I considered doing this was going wide with dataset B (making id unique)
> and merging (1:m) with dataset A, but this is annoyingly cumbersome with a lot of
> time periods and numerous variables of interest.
> Any help would be most appreciated!
> Thanks,
> Ben

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index