Sort column based on value in another column (sort rows by column values a>b>c)

I am using the function [E,index] = sortrows(A,'descend');

This sorts the first column by row value and also secondarily sorts the second column. I would then like to specify how the sort works based on relative values in a third column. For example, I am only interested in rows where column A value > column B value > column C value. I am interested in the index that identifies the rows that fit this definition. How can I set this up?

Answer 1

[Pranav Verma]

Hi Nathan,

From the question I understand that after performing the usual sorting operation on the table, you want the rows which follow the definition of "value of column A > column B > column C".

You can start by sorting the rows in whichever order you want them to be sorted (ascending / descending). After that you can use the find function in MATLAB and get the row numbers which follow the definition provided.

For eg;,

% a =  3     2     1
%      2     3     4
%      5     4     3
%      1     2     3

a = [3 2 1 ; 2 3 4 ; 5 4 3 ; 1 2 3];

% value of column A > column B > column C
b = find(a(:,1) > a(:,2) & a(:,2) > a(:,3));

% b =

%     1
%     3

Here we see that row 1 and 3 follow the given condition specified inside the find function and hence it returns the row numbers of these two rows.

Hope this helps!

Thanks

[Nathan Garcia]

Thank you! This is very helpful.