Moving sum of 4 subsequent observatins

How can I get the moving sum of 4 subsequent observations out of variable 'var'? In specific, I would like the new variable to be constructed as follows:

(81,0970 + 85,2039 + 90,0506 + 89,0082) = ...

(85,2039 + 90,0506 + 89,0082 + 100,3168) = ...

(90,0506 + 89,0082 + 100,3168 + 105,4897) = ...

...

Answer 1

[Steven Lord]

Use the movsum function.

[Joanna Przeworska]

Dear Steven,

movsum(A,[3 0]) solves my problem. Thank you.

Best regards,

JP

Answer 2

[Mathieu NOE]

hello

see example below (run it with 4 and not 5 samples averaging) :

% dummy data
data = rand(320,15);
buffer = 5;    % nb of samples for averaging 

% zero overlap mean averaging
[m,n] = size(data)

for ci=1:fix(length(data)/ buffer)
    start_index = 1+(ci-1)*buffer;
    stop_index = min(start_index+ buffer,length(data));
    time_index(ci) = round((start_index+stop_index)/2);  % time index expressed as sample unit (dt = 1 in this simulation)
    avg_data(ci,:) =mean(data(start_index:stop_index,:));  % 
end

figure(1),
plot(time_index,avg_data);

[Joanna Przeworska]

Dear Mathieu,

Thank you for your suggestion, however, to be honest, I thought there could be more simplest way to solve my problem. If am not able to find any other, I will use our idea.

Kind regards, JP

[Mathieu NOE]

hello Joanna

it's coming from long time ago when I rote most of the functions I still use today, now , as mentionned by Steven Lord below, you can do it in one line with movsum;

but basically that movsum function implements the code I show you