Vector averaging with mean command
1 view (last 30 days)
Show older comments
Hello! I don’t quite understand how to use this command, I need to average a 1x533 vector
Averaging1=mean(X,10);
Averaging2=mean(X,50);
Averaging3=mean(X,100);
I need to average over 10, 50 and 100, when I use such commands nothing happens
2 Comments
Answers (2)
Rik
on 3 Feb 2020
A faster method than a for-loop is using movmean and throwing out the results you don't need. This will also take care of the case where you have an incomplete trailing block.
clc
X=rand(28,1);
n=10;
tic
Averaging1=[mean(X(1:10)); mean(X(11:20)); mean(X(21:28))];
toc
tic
Averaging2=movmean(X,[0 n-1]);
Averaging2=Averaging2(1:n:end);
toc
isequal(Averaging1,Averaging2)
3 Comments
Rik
on 3 Feb 2020
Do you mean this should be repeated for different values of n? Or do you have a vector with all group sizes?
%example for the latter:
n=[2 4 3];
Averaging1=[mean(X(1:2));mean(X(3:6));mean(X(7:9))]
Bhaskar R
on 3 Feb 2020
Second argument of mean is dimension, your input vector X is 1x533 vector 1D vector.
1) I assume you need mean of the values from 1 to 10, 1 to 50 and 1 to 100 values
Averaging1=mean(X(1:10));
Averaging2=mean(X(1:50));
Averaging3=mean(X(1:100));
or 2) I assume you need mean of the values from 1 to 10, 11 to 50 and 51 to 100 values then
Averaging1=mean(X(1:10));
Averaging2=mean(X(11:50));
Averaging3=mean(X(51:100));
2 Comments
Rik
on 3 Feb 2020
You are ignoring the fact that your padding influences the mean value of the final block. You should probably use something like the code below.
if mod(numel(X),n)~=0
elem_count=numel(X);
X( (end+1):(n*ceil(numel(X)/n)) )=0;
X=reshape(X,n,[]);
divs=n*ones(1,size(X,2)-1);
divs(end+1)=mod(elem_count,n);
Averaging3=sum(X,1)./divs;
Averaging3=Averaging3';
else
X=reshape(X,n,[]);
Averaging3=mean(X,1);
Averaging3=Averaging3';
end
See Also
Categories
Find more on Startup and Shutdown in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!