sum special rows on a matrix with unique function

Hey guys!

what I want is to sum column 5 just if the values from row A(:,1:4) are equals as in the e.g

A = [9 1001381 18 2014 2;
    9 1002741 18 2014 45;
    17 1002452 18 2014 250;
    21 1003292 18 2014 225;  %this row is the same as the following one.
    21 1003292 18 2014 75;
    20 1002367 18 2014 24;
    27 1001579 18 2014 20;
    27 1001802 18 2014 15];

sumvtas = grpstats(A(:,5),A(:,1:4),{@sum})
[sumvtas,count]=grpstats(A(:,5),A(:,1:4),{@sum,@numel})
results=matrix(num2str(unique(A(:,1:4))),sumvtas,count, ...
              'VariableNames',{'PEx','Mat','Week','Year','sum','count'})

   This should be the solution: 

Results= 

     [9 1001381 18 2014 2;
      9 1002741 18 2014 45;
      17 1002452 18 2014 250;
      21 1003292 18 2014 300; % here's the sum it should give.
      20 1002367 18 2014 24;
      27 1001579 18 2014 20;
      27 1001802 18 2014 15];

Thanks!

Answer 1

[Walter Roberson]

A = [9 1001381 18 2014 2;
  9 1002741 18 2014 45;
  17 1002452 18 2014 250;
  21 1003292 18 2014 225;  %this row is the same as the following one.
  21 1003292 18 2014 75;
  20 1002367 18 2014 24;
  27 1001579 18 2014 20;
  27 1001802 18 2014 15];

[urows, ~, group] = unique(A(:,1:4), 'rows', 'stable');
[sumvtas, count] = grpstats( A(:,5), group, {@sum,@numel});
results = array2table( [urows, sumvtas, count], ...
            'VariableNames',{'PEx','Mat','Week','Year','sum','count'});

Answer 2

[Niko]

To clarify, do you wish to merge rows in A that have the same first four elements? If that's the case, you can try

[B, ~, idx] = unique(A(:, 1:4), 'rows', 'stable');
Results = [B, accumarray(idx, A(:, 5))];