substitute for discretize command

discretize function was available in new version of matlab..but i have older version (R2014b)in my desktop. i cannot install new version into my desktop..please help my with an alternative function for discretize in matlab

Answer 1

[Walter Roberson]

For equal intervals, separated by delta:

discrete_x = floor( (x - minimum_allowed_x) ./ delta ) .* delta + minimum_allowed_x;

For unequal intervals in which the left edges are given by the vector edges and the last entry of edges is the upper bound:

[~, ~, bin] = histcounts( x, edges );
discrete_x = edges(bin);

[Iason Grigoratos]

matlab 2014a does not have histcounts. Any alternatives?

I did the following:

data = [1 1 2 3 6 5 8 10 4 4]

binedges = 2:2:10;

[cnt, idx] = histc( data, binedges );

outside = data(idx==0) % data that fall outside the bins

binned(1) = data(idx==1) % data that fall into the first bin (2 to 4)

binned(2) = data(idx==2) % data that fall into the first bin (4 to 6) % and so on

% for data value equal to 10 is doesnt work properly though

[Walter Roberson]

[~, bin] = histc( x, edges ); 
discrete_x = edges(bin);

[Iason Grigoratos]

it does not work properly, example:

x = [1 1 2 3 6 5 8 10 4 4];

edges = 2:2:10;

[~, bin] = histc( data, edges );

discrete_x = edges(bin);

% bin takes the value of "5" when data value is "10", while the bins are 4 (N border values, N-1 bins)

-- can i just do "bin(bin>=length(edges))=length(edges)-1" ?

Also if x is outside the edges then edges(bin) returns an error, so your x must always be within the range of bins.

[Walter Roberson]

[~, discrete_x] = histc(x, edges);
discrete_x(discrete_x == length(edges)) = length(edges)-1;
discrete_x(discrete_x == 0) = NaN;

[johnson saldanha]

may i know how can i be able to get the 2nd column from the x matrix in discrete_x