Creating a variable that denotes a range of values

I have a dataset that looks like this:

Signal PT#1 PT#2 ... PT#26

0.20 2121 1232

0.21 1233 1233

0.22 1236 1232

0.23 1232 1236

...

1.00 1212 1232

...

4.00 1212 12321

I used this code to discretize the values to the hundredth's and average it for each group of values.

edges = (0:0.05:4);
allData(:,27) = discretize(allData(:,1),edges);
[C,~,IC] = unique(allData(:,27));

avg=NaN(76,26);
for i = 1:25
    avg(:,i) = accumarray(IC, allData(:, i+1), [], @mean); % calculate the mean in your data of column 2 for rows in IC with identical elements
end

Result = [C avg];

This resulted in a dataset "Result" that looks like this
1     1241    1231    1232
2     1232    1232    1231
...
76    2312    1231    1231

How do I convert the first column into ranges of the variables that were discretized? Like I want the first row to say "0.2 to 0.25" rather than "1" ?

Answer 1

[Steven Lord]

x = 4*rand(10, 1);
edges = (0:0.05:4);
y = discretize(x, edges, 'categorical');
results = table(x, y)

results = 10×2 table
       x            y     
    _______    ___________

      3.161    [3.15, 3.2)
    0.16367    [0.15, 0.2)
     1.6123    [1.6, 1.65)
     1.2792    [1.25, 1.3)
     1.7895    [1.75, 1.8)
     1.1378    [1.1, 1.15)
     1.7702    [1.75, 1.8)
      3.847    [3.8, 3.85)
      1.756    [1.75, 1.8)
     2.1521    [2.15, 2.2)

Or if you want more control over the format of the category names:

categoryNames = edges(1:end-1) + " to " + edges(2:end);
z = discretize(x, edges, 'categorical', categoryNames);
results2 = table(x, z)

results2 = 10×2 table
       x            z     
    _______    ___________

      3.161    3.15 to 3.2
    0.16367    0.15 to 0.2
     1.6123    1.6 to 1.65
     1.2792    1.25 to 1.3
     1.7895    1.75 to 1.8
     1.1378    1.1 to 1.15
     1.7702    1.75 to 1.8
      3.847    3.8 to 3.85
      1.756    1.75 to 1.8
     2.1521    2.15 to 2.2

[Abinand Rejimon]

Thank you so much! this helps a lot!