Hello list:
Thanks to Nick for recommending a running sum of matrices for my problem yesterday. I worked like a dream. Now I have a new problem.
I have created two matrices in Stata using the matrix commands. One is called "rawsum," a 438x1010 matrix. The other is "bothvoting," also a 438x1010 matrix.
I need to create matrix agreecount=(rawsum+bothvoting)/2. No problem.
Now I need to create matrix agreerate=agreecount/bothvoting, where each i,j of 'agreecount' is divided by each i,j of 'bothvoting'. I cannot find a command in Stata's matrix commands to do this type of element-wise division.
So, I've moved into Mata. Mata has a simple element-wise division command: X=A:/B.
But, I am a novice Mata-user and I cannot figure out how to get my 438x438 matrix into Mata. So, I've converted it back to a dataset using the 'svmat' command and I'm trying to get the data into Mata using the 'st_view' command. My data has 438 observations and 438 variables. Essentially, I want to use the command:
st_view(x., ., "agreecount1",..."agreecount438")
...then do the same with the 'bothvoting' data, then execute the e-w division command.
I've tried using the 'for' command but cannot figure out the successful execution of a loop in Mata because the syntax is different than Stata's.
How can I do this? Or can you provide a more elegant solution?
Thank you.
___________________________________________
Jennifer Nicoll Victor
Assistant Professor
Department of Political Science
University of Pittsburgh
4600 Wesley W. Posvar Hall
(412) 624-7204
E-mail: [email protected]
Homepage: http://www.pitt.edu/~jnvictor
*
* 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/
