Binning elements between limits

Hi all,

I have a m*2 matrix. Column 1 is the X position while Column 2 is the distance. Column 2 has a range from 0-20. I want to bin the values of column 1. Bin 1 should include all the values of X that have a corresponding distance between 0.9-1.1, Bin 2 should have all values of X having distance between 1.9-2.1 and so on. I want to have about 10 bins. I want to calculate the variance of the position elements in each bin. Is there a function to do this is MATLAB?

Answer 1

[Oleg Komarov]

Yes the function is called histc(). Read the documentation carefully.

Example:

[n,bin] = histc(A(:,2), [0.9,1.1,1.9,2.1])

Then to calculate the var():

accumarray(bin, A(:,2),[],@var)

[NS]

Thanks Oleg. I will have a look at it.

Answer 2

[John Petersen]

histc gives you the number of elements in the bins. Below code should give variance of the elements in the bins:

    % X is position
    % D is distance
    binwidth = 0.1;
    n = 10;  % number of bins
    vx = zeros(n,1);
    for k=1:n
      ind = (D>k-binwidth)&(D<k+bindwith); % index of bins
      vx(k) = var(X(ind)); % variance of X bins
    end

[NS]

Thanks John. I will have a look at it.

Answer 3

[Matt Fig]

Here I show an example of what I think you are getting at, though it is still a little unclear to me.

D = rand(30,2)*20;  % Example data.
BINS = bsxfun(@plus,[.9 1.1].',0:18);  % Example; make your own...
[I,I] = histc(D(:,2).',[-inf BINS(:).' inf]); % Pass column 2
A = accumarray(I.',D(:,1), [], @(x) {x});
A = A(2:2:end);  % Here are col 1 matches to bins of col 2.
V = cellfun(@var,A);  % And the variances of each

[NS]

Thanks Matt. I will have a look at it.