Replace elements in matrix based on indices stored in another matrix

I will probably be unable to articulate the problem well.

There are three variables, indicating indices:

a = [3:10, 17:21, 28]; b = [11:13, 22:24, 29:30]; c = [14:16, 25:27];

These are all indices in a categorical fashion (each variable is a category). Now, I want to create a variable to feed into a RepeatedMeasuresModel; hence, I will need some sort of dummy coding. To automate this, I want to have a matrix whose elements are 1, 2 or 3 (thereby representing a/b/c), based on the indices from the said variables.

E.g. conditions = [1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 1, 1, ...., 2, 2]; %n = 30. How do I output this? I have tried logical indexing, but it fails, because the matrices are not the same size. E.g. conditions = 1:30; conditions(conditions==a) = 1;

A side question, how would I, instead of say using 1/2/3 integers, use char arrays instead? For instance, instead of 1, we want to place the "value" 'Category1', 2 = 'Category 2', etc.

Thanks for your support.

Answer 1

[Star Strider]

I am not exactly certain what you want.

Try this:

a = [3:10, 17:21, 28]; 
b = [11:13, 22:24, 29:30]; 
c = [14:16, 25:27];
conditions = zeros(1,30);                               % Preallocate
conditions(a) = 1;
conditions(b) = 2;
conditions(c) = 3
cats = cell(1,30);                                      % Preallocate
cats(a) = {'Category1'};
cats(b) = {'Category2'};
cats(c) = {'Category3'}

conditions = 
    ' 0  0  1  1  1  1  1  1  1  1  2  2  2  3  3  3  1  1  1  1  1  2  2  2  3  3  3  1  2  2 '

cats =
    '    Category1  Category1  Category1  Category1  Category1  Category1  Category1  Category1  Category2  Category2  Category2  Category3  Category3  Category3  Category1  Category1  Category1  Category1  Category1  Category2  Category2  Category2  Category3  Category3  Category3  Category1  Category2  Category2  '

[LM_BU]

Hi Star, This was indeed the solution. Thank you very much!

[Star Strider]

As always, my pleasure!

Answer 2

[Stephen23]

Just use indexing, which is simple and very efficient:

>> C{1} = [3:10, 17:21, 28];
>> C{2} = [11:13, 22:24, 29:30];
>> C{3} = [14:16, 25:27];
>> cond = [1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 1, 1, 2, 2];
>> Z = C(cond)

[LM_BU]

Hi Steve,

Thank you, but this does not answer my question. C{1..3} hold indices. I want to create another variable (your "cond"), but without manually adding those values myself. The result should be something like cond = 3:30; cond(a) = 1; cond(b) = 2; cond(c) = 3; Basically, I need to "flag" a matrix of 30 columns what category each column represents, based on the indices in a, b, and c.