How to retrieve the indices of the values in each bin?

I have a histogram with values from a vector A , spread into 30 bins. How can I get the indices of the values in A, that correspond to each bin?

Example:

A = [ 4 6 8 2 5 3 3]

bin number 4 contains [4 3 3]

so I want to have a vecor B containing the indices B = [1 6 7]

Answer 1

[Walter Roberson]

The code could be slightly simpler if all of the bins were only one value wide.

A = randi(60, 1, 50)

A = 1×50

36 10 40 43 35 52 52 43 21 44 23 47 8 31 46 19 59 56 47 52 46 47 58 59 25 48 31 59 53 13

edges = [1:2:60, inf]

edges = 1×31

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59

[counts, ~, bins] = histcounts(A, edges)

counts = 1×30

1 1 0 2 1 1 1 1 0 2 2 1 3 1 1 3 0 2 0 1 1 3 2 4 1 4 2 2 4 3

bins = 1×50

18 5 20 22 18 26 26 22 11 22 12 24 4 16 23 10 30 28 24 26 23 24 29 30 13 24 16 30 27 7

B = accumarray(bins(:), reshape(1:numel(bins), [], 1), [], @(V){V.'})

B = 30×1 cell array
    {[      48]}
    {[      37]}
    {0×0 double}
    {[   13 45]}
    {[       2]}
    {[      33]}
    {[      30]}
    {[      32]}
    {0×0 double}
    {[   16 46]}
    {[    9 50]}
    {[      11]}
    {[25 31 43]}
    {[      42]}
    {[      39]}
    {[14 27 35]}

[Lu Da Silva]

but my bins are not 1 value wide.

Either way, I get an error using accumarray: "First input SUBS must contain positive integer subscripts".

and what is V?

[Walter Roberson]

If accumarray() is giving you that error, then it implies that some value in your matrix A is less than the first value in your edges vector or greater than the last one. For example,

A = randi([-2 60], 1, 50)

A = 1×50

51 59 42 31 0 39 6 59 48 56 43 30 -2 8 25 45 48 2 59 0 49 57 0 17 9 16 11 56 46 50

edges = [1:2:60, inf]

edges = 1×31

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59

[counts, ~, bins] = histcounts(A, edges)

counts = 1×30

2 0 2 1 2 1 0 3 1 0 1 1 1 0 2 1 0 1 0 2 3 1 4 3 2 3 1 3 1 3

bins = 1×50

26 30 21 16 0 20 3 30 24 28 22 15 0 4 13 23 24 1 30 0 25 29 0 9 5 8 6 28 23 25

B = accumarray(bins(:), reshape(1:numel(bins), [], 1), [], @(V){V.'})

Error using accumarray
First input SUBS must contain positive integer subscripts.

See how the 5th bin number is 0, which corresponds to the bin for the input value 0 in A, and 0 is before the first value in edges . (It is not because A has negative or 0 entries: it is strictly to do with the fact that it has entries that are outside the range of the edges list.)

[Walter Roberson]

Note that in the following code, any value in A that is outside the range of the edges will not have its index appear anywhere in B.

A = randi([-2 60], 1, 50)

A = 1×50

58 21 4 42 56 -2 4 20 34 29 40 26 58 4 60 -2 29 6 -2 44 38 28 32 33 28 52 9 1 8 33

edges = [1:2:60, inf]

edges = 1×31

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59

[counts, ~, bins] = histcounts(A, edges)

counts = 1×30

2 5 2 2 1 1 2 1 0 1 1 0 1 2 3 2 5 0 2 1 1 1 2 1 1 1 0 2 2 1

bins = 1×50

29 11 2 21 28 0 2 10 17 15 20 13 29 2 30 0 15 3 0 22 19 14 16 17 14 26 5 1 4 17

valididx = reshape(find(bins), [], 1);
B = accumarray( reshape(bins(valididx), [], 1), valididx, [], @(V){V.'})

B = 30×1 cell array
    {[       28 39]}
    {[3 7 14 31 37]}
    {[       18 38]}
    {[       29 47]}
    {[          27]}
    {[          46]}
    {[       34 50]}
    {[          42]}
    {0×0 double    }
    {[           8]}
    {[           2]}
    {0×0 double    }
    {[          12]}
    {[       22 25]}
    {[    10 17 44]}
    {[       23 40]}