Run a loop through multiple matrices

Hi, I have 15 matrices, say A1, A2 and so on.. I want to remove all the values below 0.10 and above 0.30 from all of these matrices. Then, I want to replace all of those places with NaN. Can someone please help me with this?

This might be basic, but I am very new to matlab or any kind of coding. Thanks! NK

Answer 1

[Kelly Kearney]

To do the replacement on a single variable, you just need

x(x < 0.1 | x > 0.3) = NaN;

To apply it to all your variables, there are a few possibilities.

- Type that line 15 times, replacing x with whatever the names of your 15 variables are.

- If the variables are named systematically, as you suggested (i.e. A1, A2, A3), you could try using eval, as Amit suggests. But eval is usually recommended against, because it can cause unexpected outcomes if you're not careful (and it makes for very ugly code).

- My suggestion: just create a function that applies the NaN-replacement to any number of variables.

function varargout = addnans(varargin)

for ii = 1:length(varargin)
  varargout{ii} = varargin{ii};
  varargout{ii}(varargout{ii} < 0.1 | varargout{ii} > 0.3) = NaN;
end

Then you could call it as

[A1, A2, ...] = addnans(A1, A2, ...)

This will work regardless of the number of variables, or whether the variables follow a systematic naming convention or not. Just make sure your input variables always match your output variables.

[Navneet]

Hi Kelly, That worked! Thank you so much.

NK

Answer 2

[Amit]

Something like this:

for i = 1:5
   eval(['A' num2str(i) '(' 'A' num2str(i) '<0.1 | A' num2str(i) '>0.3)' '=' 'NaN'])
end

[Navneet]

Hi Amit, Thank you for the response. Can you please explain why you chose i=1:5? Also, I actually have different names for all these matrices, so how will the loop run through all of them? Thanks!

[Navneet]

And, those names are not in the series like I had given in the example. Sorry about that!