Average values only next to each other in a matrix

I have a matrix as such: (it is much longer irl)

A = [ 62.9; 63; 83.10; 83; 43; 43; 63; 63.2; 82.9; 83; 83; 62.90; 62.90; 62.9 ];

I am looking for an efficient way to average the values that surround the values next to each that are almost identical ( +- 0.2) other as such:

A_final = [ 62.95; 83; 43; 63.1, 83; 62.90 ]

I was thinking to do it with unique, but then I risk the possibility of the same value coming up in matrix A. Is there a simple way to do so in matlab?

[the cyclist]

Step 1: Define "almost identical" mathematically.

[Jimmy Neutron]

Added it :) it is plus minus 0.2

Answer 1

[David Hill]

I don't think there is an easy way, but a simple loop works.

A = [ 62.9; 63; 83.10; 83; 43; 43; 63; 63.2; 83; 83; 62.90; 62.90 ];
A_final=[];
b=A(1);
c=1;
for k=2:length(A)
    if A(k)<1.1*b/c&&A(k)>.9*b/c
        b=b+A(k);
        c=c+1;
    else
        A_final=[A_final;b/c];
        b=A(k);
        c=1;
    end
end
A_final=[A_final;b/c];

Answer 2

[Ahmed Redissi]

You can reshape the n by 1 vector so that it is 2 by n/2 and then take the mean which will calculate the average for each column and give you the result that you want. Here is how it is done:

A = [62.9; 63; 83.10; 83; 43; 43; 63; 63.2; 83; 83; 62.90; 62.90];
n = length(A);
A_shaped = reshape(A,2,n/2);
A_final = mean(A_shaped)';