Randomising a vector with respect to the probability of the outcome

Hi, I have 60 trials which have two outcomes 0 and 1 with probabilities of 80% and 20%.

I want to randomise this 60x1 matrix in such a way that every 5 iterations, outcome 0 occurs 4 times and outcome 1 occurs once as they should.

I tried using ransrc but the 5x1 matrices that I get as an outcome very rarely match the desired probabilites.

[Mitchell Thurston]

This isn't a general case, but if I understand what you're wanting to do this'll do the job

result = zeros(60,1);
for i = 0:5:55
    result(i + randi(5)) = 1;
end

[Marek Galus]

Thanks

Answer 1

[Mohammad Alhashash]

You can use the randperm function to build 12 basic 5-by-1 matrices and then map the values into 0 and 1 based one your probabilities.

Here what do I mean:

A = [];
m = 60; %length of the main matrix A
for i=1:5:m
    a = randperm(5)';
    a(a~=1) = 0; % this based on your probability, I used 20% 1 and 80% 0
    A = [A;a];
end

[Marek Galus]

Thank you!

Answer 2

[Matt J]

I=randi(5,1,12);
J=1:12;

result=reshape( sparse(I,J,1,5,12)  ,[],1);

full(result(:))

ans = 60×1

0 0 1 0 0 1 0 0 0 0