Help with for cycle in a vector

I have a vector , and I want to obtain another vector whose elements are obtained by averaging over the elements of the first one. For example I have

x = [3,6,4,5,1,7,9,8,5]

and I want to obtain a vector by averaging every 3 elements of x, like this

y = [(3+6+4)/3,(5+1+7)/3,(9+8+5)/3].

Also I want to take into account the modulus of division, so if I have

x=[3,6,4,5,1,7,9,8]

then y should be

y = [(3+6+4)/3,(5+1+7)/3,(9+8)/2],

and if

x=[3,6,4,5,1,7,9]

then y should be

y = [(3+6+4)/3,(5+1+7)/3,9]

I know I have to use more than a 'for' cycle but don't know how, could you help me? thanks

Answer 1

[Azzi Abdelmalek]

x = [3,6,4,5,1,7,9,8,5]
p=mod(-numel(x),3)
x(end+1:end+p)=mean(x(end:-1:end-(3-p)+1))
out=mean(reshape(x,3,[]))

[aurc89]

Thanks!

Answer 2

[Andrei Bobrov]

out = blockproc(x,[1,3],@(x)mean(x.data));

or

out = nanmean(reshape([x,nan(1,mod(-numel(x),3))],[],3));