Generate N observations from a uniform distribution on [0,1] and compare it to a threshold

 We simulate this experiment by generating a logical array of
 length N, with 1 (true) for "heads" and 0 (false) for "tails". 
 The problem is to generate the right number of 1s and 0s, so that
 the percentage of 1s is p (60%). We shall do so in two ways.

 N=1000;                  // number of coin tosses
 p=0.6;                   // bias: percentage of 1s in the array

 Method 1
Generate N observations from a uniform distribution on [0,1]. 
Threshold these observations at p. The resulting logical array
will have 

 1 if an observation was < p, 
 0 otherwise.

[James Tursa]

What have you done so far? What specific problems are you having with your code? Have you looked at the doc for the rand( ) function?

[Ayman Alotaibi]

I’ll upload what I’ve done Give me few minutes plz

[Ayman Alotaibi]

percentageOfOnes = p;

numberOfOnes = (N * percentageOfOnes );

signal = [ones(1, numberOfOnes), zeros(1, N - numberOfOnes)]; signal = signal(randperm(length(signal))); numOnes = sum(signal);

[Ayman Alotaibi]

but this only gives u sum of ones, how can I have two signal with sum all ones, and sum all zeros

Answer 1

[Roger Stafford]

Here's another way. Let there be N1 trues and N0 falses.

x = [repmat(true,1,N1),repmat(false,1,N0)];
x = x(randperm(N1+N0));

[Ayman Alotaibi]

I tried urs it won’t give me the result of all 1s 60% and 0s %40

[Image Analyst]

Of course it does. Here's my proof:

N1 = 60;
N0 = 40;
x = [repmat(true,1,N1),repmat(false,1,N0)];
x = x(randperm(N1+N0));
numOnes = sum(x)  % Will say 60

Now it's your turn to give your proof that it DOES NOT work.

[Ayman Alotaibi]

can I get the sum of zeros too ? like numZeros, so it gives 40? or should I do numOnes - X ?

[Ayman Alotaibi]

I did this code at the bottom of code, and it worked

numZeros = length(zeros(1, N - numberOfOnes)) ;