How can I group data in different size categories?

Hi everyone! So I want to categorize the first column of the follwing data according to: [0;3.5;6;8;10;12;14;16;18;20;22;24;26;28;30;32;36;40]

         x         y
         0    0.0040
   17.8200    0.0030
   19.0700    0.0035
   19.1600    0.0028
   19.9300    0.0035
   20.3800    0.0208
   20.4100    0.0002
   20.4800    0.6503
   20.5100    0.0043
   20.5400    0.1609
   20.5700    0.0275
   20.6000    0.0023
   20.6400    0.0142
   20.6700    0.0025
   20.7000    0.0255
   20.7600    0.0028
   20.8300    0.0066
   20.8600    0.0032
   20.9600    0.0023
   21.0800    0.0193
   21.1200    0.0098
   21.2100    0.0112
   21.3100    0.0023
   21.3400    0.0012
   21.3700    0.0054
   21.5000    0.0029
   21.6000   15.7993
   21.6300    0.0171
   21.7600    0.0065
   21.7900    0.0034
   21.8500    3.9724
   21.9500    0.0041
   22.2000    0.0116
   22.4000    0.0025
   22.4300    1.1909

I need that in order to plot the values in bar where x will appear in categories 0-3.5, 3.5-6 etc.

Any help would be appreciated! Thank you very much!

Answer 1

[Star Strider]

Use the histc function.

I named your array xy and your bin edges binrng:

binrng = [0;3.5;6;8;10;12;14;16;18;20;22;24;26;28;30;32;36;40];
xy = [0    0.0040
   17.8200    0.0030
   19.0700    0.0035
   19.1600    0.0028
   19.9300    0.0035
         . . .      ];

counts = histc(xy(:,1),binrng)

[nath]

thank you very much for your answer!

[Star Strider]

My pleasure!

[nath]

sorry to bother again but maybe you can help. the histc function does the categorize on X as I wished but how can I have in Y axis the values of y and not just the frequency of x? I mean at the end I want a bar plot where is shown e.g. category A has value y.

[Star Strider]

I’m not certain that what you want is possible. The closest I can come to it is to use a bar3 plot:

[counts, idx] = histc(xy(:,1),binrng);

figure(1)
bar3(xy(idx,1), xy(idx,2))

These are frequency counts, so there are restrictions on how you can plot them.

[nath]

you are right, it is not possible, i will probably have to sum the values of y for each bin.

thanks anyway!!

[Star Strider]

My pleasure!

The bin distributions for y are referred to with the idx vector that histc returns. They will therefore have the same bin counts as x because you are using x to categorise them. If you bin the y values using your current bin assignments, all will be assigned to the first bin, except for one (15.8) that will be assigned to the eighth.