How to calculate moving average for data

How could I calculate moving average for specified data , having specific overllaping (For example I have a set of data, and I would like to compute moving average). How to make it?

Also I do noy understand how to use movmean(A,k)? What should I put in k value in this function?

Could you please help me?

[KALYAN ACHARJYA]

Please share the data example (sample)

[Ivan Mich]

I am uploading the file. It is just a sample of full data I have

[Rik]

What do you mean with 40% overlap?

[Ivan Mich]

Nevermind about overlap. Could you please help me about calculating moving average?

[Rik]

What exactly is it you want to do? For each nth element of your vector you can specify which range of values should be averaged. As the documentation for movmean explains, k denotes how wide that window should be.

Answer 1

[KALYAN ACHARJYA]

Here One way: Using Loop (As the data is 1 D)

data=1:60 % Random data as per Attached sample
% Lets Decided the Segment size 5 Data Elements
% Move the window by 3 Elements (60% of 5, remaining 2 (40%) Overlap)
move_av=zeros(1,length(data)/3);
idx=1:3:length(data);
for i=1:length(idx)-4
    move_av(i)=mean(data(idx(i):idx(i)+4));
end 
move_av

Other related function to avoid the loop blockproc, I will respond as a comment once I do. Or you may get the respond from the expert members

https://in.mathworks.com/help/images/ref/blockproc.html

[Ivan Mich]

Thank you, but I want to ask a question. I do not understand the line

for i=1:length(idx)-4

Why are you subtract the value 4? what is 4 ?

And could you explain me the line?

    idx=1:3:length(data);

[KALYAN ACHARJYA]

If you consider Block Segemnt is 5, you have consider 5 indices for average

Lets say present index 6, the moving average should be 6 to 10 (5 Data Elements)

Hece 6:6+4

>> 6:6+4
ans =
     6     7     8     9    10

Next Question Why -4: The loop iterate till the 4 elements from the end, as you have compensate to avoid array exceeds matrix dimension. Please take a pen and paper, you may understand it easily.

[Ivan Mich]

Ok, and why segment size is 5 ? could you please explain it?

[KALYAN ACHARJYA]

It's an Example, you may consider any other window size for moving average. Note you have to adjust the moving indices as well

Once is window size : Here 5

Moving window by 3, so that 2 data elements (40% of 5) would be ovelap

[Ivan Mich]

ok. If I have 4000 data numbers and I want window size equal to 1 how should I modify the code? movmean function is wrong to be used?

[Ivan Mich]

I am uploading the file

[KALYAN ACHARJYA]

If you consider the moving window size is 1, then its simply average of individual data elements divided by 1. And the result will be same as data

>> data=1:10
data =
     1     2     3     4     5     6     7     8     9    10
     

Here

>> movmean(data,1)
ans =
     1     2     3     4     5     6     7     8     9    10

More:

movmean(data,k)
Here k represents the window size 

Lets consider k is 3

Result

>> movmean(data,3)
ans =
    1.5000    2.0000    3.0000    4.0000    5.0000    6.0000    7.0000    8.0000    9.0000    9.5000