concatenation of few matrix with a loop

C{1} = [total1979_125_30];

C{2} = [total1980_125_30];

C{3} = [total1981_125_30];

C{4} = [total1982_125_30];

C{5} = [total1983_125_30];

...upto C{40} = [total2019_125_30];

just help me making the loop

So that I can use

A = cat(3,C{:});

M = mean(A,3);

[Stephen23]

See original question here:

https://www.mathworks.com/matlabcentral/answers/606111-averaging-every-element-of-some-matrices

[Stephen23]

The cause of your problem is that you put meta-data (e.g. dates) into variables names. Meta-data is data and belongs in variables, not in variable names. Once you have made that bad design decision, you force yourself into writing slow, complex, inefficient, obfuscated, buggy code to try and access those multiple data arrays:

https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

So far you have not told us the most important information: how did you get those arrays into the workspace? Did you write them all out by hand? My guess is that you used load in a loop, in which case you can fix the problem at its cause by loading into an output variable, e.g.:

N = the number of files
C = cell(1,N);
for k = 1:N
    F = the k-th filename
    S = load(F);
    C(k) = struct2cell(S);
end

Note how this does not require evil eval anywhere. Once you tell us how you actually get those multiple variables into the workspace, someone can help you to import/create the data in a better way.

Answer 1

[Ameer Hamza]

It was not a good idea to create a variable name like this: https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval. Always use a cell array. Since you already created these, you can use the following loop to create the cell array now.

C = {};
vals = 1979:2019;
for i = 1:numel(vals)
    [~, C{i}] = evalc(sprintf('total%d_125_30', vals(i)));
end

[Joydeb Saha]

how can I deduct a specific matrix from all other 40 matrices using a loop?

as like...

C1 = (total1979_125_30-M);

C2 = (total1980_125_30-M);

C3 = (total1981_125_30-M);

.

.

.

C40 = (total2019_125_30-M);

[Ameer Hamza]

What dou mean by deduct?

[Joydeb Saha]

Subtraction

[Ameer Hamza]

You can use cellfun(). For example, if your matrix name is M and cell array is C

M % matrix to subtract
C % 1x40 cell array of matrices
C_new = cellfun(@(x) x-M, C, 'uni', 0);

[Stephen23]

"how can I deduct [subtract] a specific matrix from all other 40 matrices using a loop?"

If you stored all of the data in one 3D array, then with one simple subtraction.

[Ameer Hamza]

Joydeb Saha comment posted as answer moved here

Ameer Hamza its showing the error : Undefined operator '-' for input arguments of type 'cell'.

the cell array C is in 1x41 cell form and M is in 12x1 double form

[Ameer Hamza]

Can you attach the variables C and M as a .mat file?

[Joydeb Saha]

Yes I have attached.

[Stephen23]

Fix this typo:

C_new = cellfun(@(x) x-M, C, 'uni', 0);
%                 ^  ^                    must be the same !

[Ameer Hamza]

Jaydeb, as Stephen mentioned, there was a typo. The corrected line works fine.

[Ameer Hamza]

Or the alternate solution suggested by Stephen to use a 3D array

C = cat(3, C{:});
M_new = C - M;

M_new will also be a 3D array.

[Joydeb Saha]

Yes, its working. Thank you very much for both your support.