turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-02-2012 12:31 AM

A simple module for finding row-wise cumulative sum in IML:

Proc IML;

Start rowcumsum(x);

Step1 = cusum(x);

Step2 = 0 // Step1[1:nrow(Step1)-1,ncol(Step1)];

Output = Step1 - repeat(Step2,1,ncol(Step1));

return(Output);

Finish rowcumsum;

a = {1 2 3,4 5 6, 7 8 9};

SASCUSUM = cusum(a);

ROWCUSUM = rowcumsum(a);

print a; print SASCUSUM; Print ROWCUSUM;

Quit;

a

01 02 03

04 05 06

07 08 09

SASCUSUM

01 03 06

10 15 21

28 36 45

ROWCUSUM

01 03 06

04 09 15

07 15 24

Accepted Solutions

Solution

07-11-2017
09:14 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-02-2012 09:12 AM

You didn't ask a question, so I can't tell whether you are saying "here is how to do this" or whether you are asking "is this the best way to do this."

If the former, yes, this is a good way to compute cumulative sums across rows of a matrix.It is a clever algorithm.

If you are asking whether this is the fastest way, I'll offer two observations: you can use the LAG function (introduced in SAS/IML 9.22) to speed up Step2, and you don't need the REPEAT function when forming the OUTPUT matrix. Thus the following code is slightly more efficient:

start rs(x);

Step1 = cusum(x);

Step2 = lag(Step1[,ncol(Step1)], 1); /* returns missing in Step2[1] */

Step2[1] = 0; /* replace missing value with 0 */

Output = Step1 - Step2;

return( Output);

finish;

Unless your matrices are huge, it doesn't really matter which module you run. In fact, unless the number of rows is huge, you can just use a simple DO loop and the computation is just as fast:

Output = j(nrow(x), ncol(x));

do i = 1 to nrow(x);

Output[i,] = cusum(x[i,]);

end;

All Replies

Solution

07-11-2017
09:14 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-02-2012 09:12 AM

You didn't ask a question, so I can't tell whether you are saying "here is how to do this" or whether you are asking "is this the best way to do this."

If the former, yes, this is a good way to compute cumulative sums across rows of a matrix.It is a clever algorithm.

If you are asking whether this is the fastest way, I'll offer two observations: you can use the LAG function (introduced in SAS/IML 9.22) to speed up Step2, and you don't need the REPEAT function when forming the OUTPUT matrix. Thus the following code is slightly more efficient:

start rs(x);

Step1 = cusum(x);

Step2 = lag(Step1[,ncol(Step1)], 1); /* returns missing in Step2[1] */

Step2[1] = 0; /* replace missing value with 0 */

Output = Step1 - Step2;

return( Output);

finish;

Unless your matrices are huge, it doesn't really matter which module you run. In fact, unless the number of rows is huge, you can just use a simple DO loop and the computation is just as fast:

Output = j(nrow(x), ncol(x));

do i = 1 to nrow(x);

Output[i,] = cusum(x[i,]);

end;

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-03-2012 06:15 AM

Thanks for the variations, Rick. I just wanted to share the piece of code that I repeatedly found useful.

My co's yet to make the step up from 9.20, which is slightly irritating because 9.22 has a decent set of functions introduced ... so instead of say, cuprod(x) which is available in 9.22, I use exp(cusum(log(x)))

The loop increases the computation time drastically if this function is used within an optimization code. I noticed IML has this amazing speed-up (nearly 15x) for vectorized vs loop section (similar results to your blog post on running mean and variance).The results were a bit surprising because MATLAB hardly produced a 1.5x speedup for the same optimization code, and a fully vectorized optimization code in IML runs 3x faster than the same code in MATLAB!

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-03-2012 06:32 AM

Impressive results! Congratulations, and thanks for sharing!