Is there a better way to perform a moving window of operations without using a for loop?

Say I have a matrix:

x=1:100;

And I want to create a new matrix y that has the following output:

[sqrt(1^2+4^2), sqrt(2^2+5^2), sqrt(3^2+6^2), ... etc.]

Note that each term in the output is y[index]=sqrt(x[index]^2+x[index+3]^2).

I could accomplish this with a for loop, but if matrix x is really large this is inefficient. Is there a more efficient way?

Answer 1

[John Chilleri]

Hello,

You can figure out a way to use vector notation,

x = 1:100;
y = sqrt(x(1:end-3).^2+x(4:end).^2);

I ran this vs the for loop version you have above, and this performed twice as fast when x = 1:1000000 and three times faster when x = 1:100000000. Although it didn't take long for either, I'd be more concerned for memory than speed - but that's only regarding this problem.

Hope this helps!

[razorfin]

Thanks, this is exactly what I was looking for, but I just couldn't seem to think in this way. I'll remember it for the future!

[John Chilleri]

Glad I could help!