# How do I create array of a function with for loop?

6 views (last 30 days)
Jooris Preiser on 7 Sep 2021
Commented: Stephen on 10 Sep 2021
Hi, I am trying to find a shorter and easier for my calculations. I have a in total 42 of 424x412 doubles for which I want to calculate the mean of all elements and build an array of the results that would be 42x1. So far I have done that the following way
B = [mean(X1,'all');mean(X2,'all');..mean(X42,'all')]
Since I have to calculate such arrays multiple times (I have do that for multiple sets of 42 arrays) I would much rather like to that in a for-loop..so far i have tried the following, which didn't work
for k = 1:42
B = mean('X(k)','all')
end;
can somebody help me?
I am a MATLAB rookie! :D
Stephen on 7 Sep 2021
You did not tell us the most important information: how did you get 42 numbered variables into the workspace?
Did you sit and write out their names by hand? Or generate them in a loop? Or load them from file?
In any case, that is where you need to fix your code, rather than trying to fix the problem later:

Steven Lord on 7 Sep 2021
I have a in total 42 of 424x412 doubles
Can you do this? Yes.
Should you do this? The general consensus is no.
Rather than having 42 individual variables why not make a 42-by-424-by-412 array?
##### 2 CommentsShowHide 1 older comment
Stephen on 7 Sep 2021
"And I also have to do that multiple times... The X in the example above stands for a month... I want to calculate that for all 12 months and over muliptle datasets "
Sure. That would be easy and very efficient using standard MATLAB coding, e.g. using indexing to access data in arrays (which could be container arrays). Is there any particular reason why you want to force yourself into writing slow, complex, obfuscated, inefficient code?

Mathieu NOE on 7 Sep 2021
hello
it's not the best coding practice, but it works ...
clc
clearvars
% dummy data
X1 = rand([424,412]);
X2 = rand([424,412]);
X3 = rand([424,412]);
% main code
for k = 1:3
B(k) = mean(eval(['X' num2str(k)]),'all');
end;
plot(B,'*')
Stephen on 10 Sep 2021
Note that this approach will not help you, it will hinder you.