Building a matrix

[hwangnyc]

Hi everyone,

Here is a sample of my data.

The OBS is a unique individual.

Each column is a state that they have visited. 1=yes, 0=no

I'd like to to figure out the various combinations of states.

For example: OBS 1 and 7  has visited NewYork and Texas so NewYorkTexas would be 2

I'd like to do that for all possible combinations. Any suggestion as to how to begin?

OBS	NewYork	California	Texas	NewJersey	Washington
1	1	0	1	0	0
2	0	1	1	0	0
3	0	0	1	0	1
4	1	1	0	0	1
5	0	0	0	0	1
6	0	1	1	0	0
7	1	0	1	0	1
8	1	1	0	1	1
9	1	1	1	0	0
10	1	1	0	1	1

[Astounding · Accepted solution]

An easy way would be to string together all the flags:

all_locations = cat(NewYork, California, Texas, NewJersey, Washington);

You need to preserve this logic, so you know which position represents which state.

Then count:

proc freq data=have;

tables all_locations;

run;

[hwangnyc]

That was a astounding answer! THANK YOU!

[Andygray]

Hi @hwangnyc Can you post the output for the sample you provided for the benefit of the wider audience. Thank you! 

[Reeza]

Run PROC CORR with the SSCP MATRIX.
https://gist.github.com/statgeek/a5184a4e1678d81e2643#file-sas_distance_matrix-sas

This will give you a matrix of the trips.