Grouping Data Into Smaller "Bins"

Hi everyone,

I have a matrix G containing a number of data points (say 5000 for the sake of this example). I would like to create a new matrix, K consisting of only 50 elements, so that every element is the average of the corresponding 100 elements in matrix G.

For example the third element of K should be the mean of elements 100 to 150 of G.

Any way to quickly do that?

Answer 1

[Jan]

G   = rand(5000, 3);  % "data points" is not clear
N   = 100;
siz = size(G);
GG  = reshape(G, [N, siz(1)/N, siz(2)]);
R   = reshape(sum(GG, 1) / N, siz(1)/N, siz(2));

If the length of the first dimension is not a multiple of N, you might want to crop the input at first.

See also https://www.mathworks.com/matlabcentral/fileexchange/24812-blockmean:

R = BlockMean(G, N, 1) / N;