Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

To resolve issues starting MATLAB on Mac OS X 10.10 (Yosemite) visit: http://www.mathworks.com/matlabcentral/answers/159016

summation inside a "for loop"

Asked by Nilanjan Roy on 29 May 2012

Hi I have a 100x100 matrix, A. I am doing the following:

 q=0:0.1:50;sym k;
for j=1:length(q)
        B1(j)=sum(exp(q(j)*A(k,1)), k=1..100);
end
for j=1:length(q)
        B2(j)=sum(exp(q(j)*A(k,2)), k=1..100);
end
..................................
..................................
for j=1:length(q)
        B100(j)=sum(exp(q(j)*A(k,100)), k=1..100);
end

The problem is perhaps with the command but I tried with both sum/symsum and some other variations. But always matlab gives one or the other error.

It would be very helpful if someone could suggest some "correct" command. Thanks a ton.

1 Comment

Walter Roberson on 29 May 2012

http://matlab.wikia.com/wiki/FAQ#How_can_I_create_variables_A1.2C_A2.2C....2CA10_in_a_loop.3F

Nilanjan Roy

Products

No products are associated with this question.

1 Answer

Answer by Geoff on 29 May 2012
Accepted answer

I would expect this to work:

B = zeros(length(q), size(A,2));
for j = 1:length(q)
  B(j,:) = sum( exp(q(j)*A) );
end

3 Comments

Geoff on 29 May 2012

Oh, sorry, I didn't pick up that this is the symbolic toolbox... I have no idea about that. I thought that looked like strange MatLab code. I'll leave my answer just in case it's any use.

Nilanjan Roy on 29 May 2012

thanks a lot man. i think i was unnecessarily using the symsum command where all i needed was "sum". thanks again.

Geoff on 29 May 2012

Cool. Hope it's clear what's happening... sum will work along a single dimension. By default this is 1. That means it will sum columns. If it's 2, it'll sum rows, and so on.. Whatever you call strips in higher dimensions =) So we can compute the sum for every column at once (giving a 1x100 vector) and then shove that into the matching columns of B for a single row indexed by j. It would probably be more efficient if you flipped the dimensions of B and wrote a whole column at a time, but I kept it consistent with your matrix A for clarity.

Geoff

Contact us