Summing with respect to values within variable instead of rows, columns, or equal windows

I have a variable with a similar structure to the following but with many more rows:

a=
1  0.3
1  0.1
2  0.7
2  0.6
2  0.3
2  0.9
3  0.5
4  0.8
4  0.1
5  0.1
5  0.9
5  0.4

Say I want sums of column 2 with respect to each scalar (eg. 1, 2...5) so that I would get a new variable like this:

aa=

1  0.4
2  2.5
3  0.5
4  0.9
5  1.4

I have not found how to sum anything but rows and columns, so how should I go about summing based on values within a variable?

Answer 1

[Sean de Wolski]

a=[
1  0.3
1  0.1
2  0.7
2  0.6
2  0.3
2  0.9
3  0.5
4  0.8
4  0.1
5  0.1
5  0.9
5  0.4]

uv = unique(a(:,1),'stable');
sidx = accumarray(a(:,1),a(:,2));
[uv sidx]

accumarray() to the rescue!

Answer 2

[Azzi Abdelmalek]

v=unique(a(:,1),'stable');
s=zeros(numel(v),1);
for k=1:numel(v)
  s(k,1)=sum(a(ismember(a(:,1),v(k)),2));
end
out=[v s]

%or without a loop

v=unique(a(:,1),'stable');
out=[v arrayfun(@(x) sum(a(ismember(a(:,1),x),2)),v)];