MATLAB: Choose a number from a matrix based on probability parameter

MATLABprobability rndscr

I have two different procedures to do, precedure A has 0.8 probabilty and B has 0.2. How can I create a parameter to select those procedures based upon them probability in case of iterative process ( I am doing this thousands of times)?

I have seen some answers here but not so sure that I am doing it right. Two solution are below but I got suspisous its correct when run it a multiple of 10 times where the 1's or 2's werent exactly showing at 80%,20% everytime for the first solution and the use of 'true' in the second one .

I am not expert instastics and probabilty so just asking if that correct way of doing it and is the result ok or not .

% solution 1
out = randsrc(1,1,[1,2;0.8,0.2])
if out==1 
    Procedure A;
else
    procedure B
end
% solution2 
p=[1 2];
y = randsample(p,1,true,[0.8,0.2])

Best Answer

Either are overly complicated:

if rand > 0.8
    do_ProcedureA;
else
    do_ProcedureB;
end

Related Solutions

MATLAB: How can i create a sequence that every symbol has the same probability as the others

Actually, you are wrong in this. randn produces NORMALLY distibuted random numbers.

normcdf(0.5)
ans =
      0.69146

So the sequence you generated has 1 appearing almost 70% of the time. Admittedly, sometimes I type randn when I intended to use rand.

N = 1e6;
Seq = randn(1,N) < 0.5;
hist(Seq)

If you wanted to use randn, you needed to compare the elements to ZERO, not 0.5.

Seq = randn(1,N)<0;

You could also have used rand that way, by comparing them to 0.5. Thus:

Seq = rand(1,N)<0.5;

A random sequence with 0,1,2,3 with equal probability is easy enough.

Seq = floor(rand(1,N)*4);

Or you could use randi.

randi([0,3],[1,N])

Finally, each of the above schemes will generate equi-probable sequences of elements, but there is no assurance you will get EXACTLY the same number of events in each bin. That only happens over the long term, where the events will tend to become approximately equal in count. In fact, it is unlikely that you will get EXACTLY equal numbers of events.

For example, given a sequence of 1000 events, what is the probability you would see EXACTLY 500 events in each bin from a true random sampling? This is something we can compute from a binomial distribution.

nchoosek(1000,500)/2^1000
ans =
     0.025225

So only 2.5% of the time would we expect an exactly equal count of 500 heads and 500 tails from a fair coin tossed 1000 times.

Anyway, if that is what you want, then Image Analyst has shown you how to do that.

MATLAB: Generate a random signal in Matlab

y = (rand(10000,1) > 0.5)*2 -1;

Best Answer

Related Solutions

MATLAB: How can i create a sequence that every symbol has the same probability as the others

MATLAB: Generate a random signal in Matlab

Related Question