# compute sums by accumulating in a for-loop

1 view (last 30 days)
Kyle on 3 Dec 2023
Commented: Kyle on 3 Dec 2023
I am trying to accumulate sums (phi(l) for lags " l"in a for loop given the code below. I keep getting "Unable to perform assignment because the left and right sides have a different number of elements." I have tried multiple ways to try and fix it but nothing is working.
x=normrnd(0,1,100,1);
xprime=detrend(x,'constant');
phi=zeros(14,1);
L= [0:13]';
for l=1:14
phi(l)=(sum(xprime(1:100-l)).*xprime(1+l:100)/(100-(l-1)-1));
end
Unable to perform assignment because the left and right sides have a different number of elements.

Image Analyst on 3 Dec 2023
Look at this:
x=normrnd(0,1,100,1);
xprime=detrend(x,'constant');
phi=zeros(14,1);
L= [0:13]';
for l=1:14
temp = (sum(xprime(1:100-l)).*xprime(1+l:100)/(100-(l-1)-1))
fprintf('The size of temp = %d.\n', numel(temp))
phi(l)=(sum(xprime(1:100-l)).*xprime(1+l:100)/(100-(l-1)-1));
end
temp = 99×1
0.0116 0.0105 0.0110 0.0077 -0.0046 -0.0012 -0.0141 0.0183 0.0027 -0.0052
The size of temp = 99.
Unable to perform assignment because the left and right sides have a different number of elements.
So you're trying to stuff 99 values into a slot meant for only one value, phi(l). Not sure how to fix it because I'm not sure what your intent is. Did you mean for phi to be a 99 by 14 matrix and you want to put temp into the columns of phi?
##### 3 CommentsShow 1 older commentHide 1 older comment
Image Analyst on 3 Dec 2023
Watch your parentheses. Maybe you meant either
phi(l) = sum(xprime(1:(100-l)) .* xprime(l:100) / (100-(l-1)-1));
or
phi(l) = sum(xprime(1:(100-l) .* xprime(l:100)) / (100-(l-1)-1);
And l (ell) is not a good variable name - it looks too much like 1 (one) and I (capital I). Use k instead.
Kyle on 3 Dec 2023
This worked thank you.

Torsten on 3 Dec 2023
Edited: Torsten on 3 Dec 2023
sum(xprime(1:100-l))
This is a scalar.
sum(xprime(1:100-l)).*xprime(1+l:100)
This is a vector of length 100-(1+l)+1.
(sum(xprime(1:100-l)).*xprime(1+l:100)/(100-(l-1)-1))
This remains a vector of length 100-(1+l)+1.
phi(l)
This is s acalar.
Thus you try to assign a vector to a scalar which is not possible.
Maybe you mean
phi(l)=sum(xprime(1:100-l).*xprime(1+l:100))/(100-(l-1)-1));

### Categories

Find more on Correlation and Convolution in Help Center and File Exchange

### Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!