Question

I have two column vectors [a b] and [c d]. I want to select only those data points in ‘a’where ‘c’ matches to ‘a’ and also corresponding ‘b’ values. As there is some repetition in ‘a’ which is making ‘ismember’ command not to work properly. ‘For loop’ is another option but size of column vectors is too big for making a loop. Just short example is given below. Any solution? Thanks

[a b]

ans =

1.0000    0.1000
2.0000    0.2000
2.0000    0.2000
3.0000    0.4000
4.0000    0.5000

[c d]

ans =

 2     7
 3    12

Answer 1

If repeated values for a always come with the same value for b, you can use the second output of ismember for a simple solution:

ab=[1.0000    0.1000
2.0000    0.2000
2.0000    0.2000
3.0000    0.4000
4.0000    0.5000];
cd=[2     7
3    12];


[~,idx]=ismember(cd(:,1),ab(:,1))
out = ab(idx,:)

out =

    2.0000    0.2000
    3.0000    0.4000

If repeated values for a sometimes can have different values for b (e.g. [2 0.2; 2 0.3], you need to call unique on a first:

[~,idx] = unique(ab(:,1));
ab = ab(idx,:);
out = ab(ismember(ab(:,1),cd(:,1)),:);