Efficient Way To Create Adjacency Matrix in Stata?

Question

I have data that looks like this in Stata:

Cat#1    Cat#2
  A        X
  B        X
  A        Y
  C        Z
 ...      ...

I want to create an adjacency that I will later output to Matlab for further analysis.

Here is my proposed way of doing it:

• Create a column for each of the unique elements in Cat#2 (i.e. columns: X Y Z in the above example).
• Collapse across CAT#1 so that only one row for each unique Cat#1 value (i.e. rows: A B C in the example above).
• Replace each column element of X Y Z ... by 1 IF the "entry" > 0.
• Drop excess stuff, left with adjacency (matrix) dataset.

The problem is that I have a lot of unique categories for Cat#2, so I think that coding up this part would take a long time. Are there any suggestions on what I can do to speed up the process of creating the sought data structure?

Thanks!

Answer 1

There are a few Stata commands for making adjacency matrices.

Try typing findit netsis or findit nwcommands at the command prompt.