Extracting the mean of 2nd column based on unique 1st column values.

I have this vector,

1 22.6391

1 22.6357

1 22.6338

1.5 22.1159

1.5 22.1056

1.5 22.0950

1.5 22.08820

2 22.11870

2 22.11100

2 22.10280

2 22.09650

And I want a mean of the 2nd column values based on the unique 1st column values. Like,

1 22.63

1.5 22.10

2 22.10

I have tried using tmp = [unique( SINR_tmp(:,1) ),accumarray( SINR_tmp(:,1), SINR_tmp(:,2), [], @mean )];

but this gives unique values only when 1st column has integer values. But I need to include the floating values too in the 1st column.

I have also tried using the function unique() in various ways. But could not succeed in getting the desired result.

I would really appreciate any kind of help on this.

Thank you,

Rahul Singh Gulia

Answer 1

[Voss]

Here are two similar methods that produce identical results:

SINR_tmp = [ ...
    1	22.6391; ...
    1	22.6357; ...
    1	22.6338; ...
    1.5	22.1159; ...
    1.5	22.1056; ...
    1.5	22.0950; ...
    1.5	22.08820; ...
    2	22.11870; ...
    2	22.11100; ...
    2	22.10280; ...
    2	22.09650; ...
    ];

% Method 1:
uS = unique(SINR_tmp(:,1));
uS(:,2) = NaN;
for ii = 1:size(uS,1)
    uS(ii,2) = mean(SINR_tmp(SINR_tmp(:,1) == uS(ii),2));
end
disp(uS);

    1.0000   22.6362
    1.5000   22.1012
    2.0000   22.1073

% Method 2:
[uS,~,jj] = unique(SINR_tmp(:,1));
uS(:,2) = NaN;
for ii = 1:size(uS,1)
    uS(ii,2) = mean(SINR_tmp(jj == ii,2));
end
disp(uS);

    1.0000   22.6362
    1.5000   22.1012
    2.0000   22.1073

[Rahul Gulia]

Thank you for the quick reply Benjamin. I learnt a new thing today.