Cumulative mean over constantly increasing interval

Hey guys, any help would be much appreciated as I'm relatively new to the program!

I've got 400000 odd rows of data and what i need to do is calculate the mean over the first 10000 points then over the first 20000 then 30000 until I've used all the data points.

The code I've written calculates the mean over each of the 400, 10000 point intervals but doesnt increase the interval size by 10000 from the original each time.

Thanks in advance

intervalLength = 10000;

x = A.data(1:4000001,2);

%create index array idx = zeros(length(x),1); idx(1:intervalLength:end) = 1; idx = cumsum(idx);

avg = accumarray(idx,x,[],@mean);

out = avg(idx);

m = avg(1:401,1); n = [1:401]; plot(n,m); xlabel('interval number'); ylabel('x_average');

Answer 1

[Star Strider]

I’m not sure what you’re doing. so I’ve done two options:

D = rand(4E+5, 2);                              % Create Data

r = [1E+4 : 1E+4 : size(D,1)];
for k1 = 1:length(r)
    m1(k1,:) = mean(D(1:r(k1),:));              % Takes ‘mean’ Of [1:1E4], [1:2E4], &c.
end

DR = reshape(D, 1E+4, size(D,2), []);
m2 = squeeze(mean(DR, 1)).';                    % Takes ‘mean’ Of All 40 1E+4 Row Sections Individually

I doubt accumarray is going to do what you want, at least any more efficiently that what I’ve done here (unless I’m not understanding what you want to do).

[Yusuf]

Thanks Star Strider, first option is exactly what I was after!

[Star Strider]

My pleasure!