Consider only a positive numbers of the first columns.

I have this matrix:

A =

   -0.0001   -0.0723    0.0007    0.0013   -0.0007   -0.0016   -0.0009    0.0010   -0.0018
   -0.0004   -0.0723    0.0008    0.0012   -0.0008   -0.0017   -0.0009    0.0011   -0.0017
   -0.0005   -0.0723    0.0007    0.0013   -0.0008   -0.0017   -0.0009    0.0011   -0.0017
    0.0002   -0.0723    0.0007    0.0012   -0.0007   -0.0017   -0.0009    0.0010   -0.0018
   -0.0003   -0.0723    0.0006    0.0012   -0.0007   -0.0017   -0.0009    0.0011   -0.0017
    0.0005   -0.0723    0.0007    0.0013   -0.0007   -0.0017   -0.0009    0.0011   -0.0017

My issue is to by-pass (or not consider) the rows that start with a negative number in the first column. How can I extend this procedure if I work with cell or struct variable? How can I do it?

Answer 1

[Mischa Kim]

Use

 B = A(find(A(:,1)>0),:)

Concerning your second question: that, of course, depends on how your cell array looks like. In general, you could convert to a numeric array and then use the code above.

[Francesco]

The general configuration is this one:

A = 

    Columns 1 through 4

      [1000x9 double]    [1000x9 double]    [1000x9 double]    [1000x9 double]

    Columns 5 through 8

      [1000x9 double]    [1000x9 double]    [1000x9 double]    [1000x9 double]

Are you sure that it's a good strategy to convert to a numeric array and then use the code abbove?

[Francesco]

Sorry Kim, but with your code MATLAB said:

 Empty matrix: 0-by-9

What' the matter?

[Francesco]

Sorry Kim, my mistake! :D

[Mischa Kim]

Well, as I said, it depends on how your cell array looks like. So for the one you're posting in your comment: do you want to treat the 8 different arrays as independent ones with regards to ignoring rows that start with a negative number in the first column? Or is this supposed to be one large 1000-by-72 matrix?

[Francesco]

It's not supposed to be one marge 1000 by 72 matrix. But I would like to to the same procedure you said for each one of the 8 (or n) different arrays. As you said I want to: "treat the 8 different arrays as independent ones with regards to ignoring rows that start with a negative number in the first column"