Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: Creating square matrix with obs counts from 2 non unique id variables


From   Diogo L Pinheiro <diogo.pinheiro@pubpolicy.gatech.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Creating square matrix with obs counts from 2 non unique id variables
Date   Tue, 19 Apr 2011 11:47:34 -0400

Hello, everyone,
I've been trying to create a square matrix with counts of observations from two non unique ID variables. In this case, one id variable has the code for an organization an individual was and one for where the individual is. For example:


Observation    ID1         ID2
1.                  10          12
2.                  10          11
3.                  11          12
4.                  11          10

In the example above, observation 1 was at organization #10 at one time and organization #12 at another. Observation 2 was at 10 and then 11. And so on.


From that, I want to create a square matrix that would look something like:


              10           11           12

10           0              1            1
11           1              0            1
12           0              0            0

So that I could have all the counts of each combination of organizations.

The main problem is getting a square matrix in the end. I can create a non square matrix in this case by using
tab ID2, gen(newID)
collapse (sum) newID, by(ID1)

but the problem is that this won't be a square matrix, as not every ID number shows up in both variables.. In the example above it would be missing a line (since not every ID number present in ID1 is present in ID2, and vice versa).

So how could I get a square matrix from the situation above? I've found commands to create similar matrices, but they require unique IDs and only provide binary results for the existing combinations.


Thanks!
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index