Compute a new vector dependent on two others and a random probability

Hi everyone, I have the following problem which I can not figure out by myself. I have two column vectors (x &y: 200x1 each). Each entry is giving me percentages (0.2 0.4 or 0.6). I want to compute two vectors (z,k) based on the old vectors percentages. So if either x or y == 0.2, I want z or k to be 1 in 0.2% of occurences in the old vector and 0 in all the other cases. It does not matter whether z or k = 1.

Here is what I have tried so far:

if matrix(:,1)==0.2|matrix(:,2)==0.2
  % matrix(:,5)==1
    matrix(:,6)==1
end

Here I only wanted to try to find out the occurences where 0.2 is either in column 1 or 2 and then randomly delete the 1 entrys from column 5 and 6. But this already did not work :(.

Please help me!

[KSSV]

Procedure seems fine enough..have you initialized the matrix?

[Laurie Gerstenberger]

Thanks for your answer.yes, I did.

[KSSV]

Are you facing a problem to find out where 0.2 is? As the numbers are floating try abs(matrix(:,1) -0.2)<eps where eps is a small number.

Answer 1

[Massimo Zanetti]

By doing

x==0.2 | y==0.2

you are testing that ALL theelements of x or y equal 0.2. Is that really what you want?

[Laurie Gerstenberger]

true..Thanks! No I want that either one element in a row==0.2

[Laurie Gerstenberger]

I solved that now and now I have the position where either on of them is 0.2. Is there any way to write 1 randomly in either vector z or y randonly and only in 20% of the occurences of the positions?