0

I am attempting to create a gamma distribution in MATLAB; however, I keep receiving the error:

Index in Position 1 exceeds array bounds (must not exceed 100).

Assuming I am reading this correctly, it is referring to variable M that is simply = 2500 (the number of pseudo-random variables I am using for this project).

I was hoping someone can explain what is wrong with my logic and possibly a solution.

alpha = 0.5;
w = gamma_rdn(M,alpha);
x1 = (0.0001:0.001:1); % For plot

figure(5)
subplot(2,1,1);hist(w);title('Histogram of Gamma RDN');
subplot(2,1,2);plot(x1,pdf('gam',x1,alpha,1));title('Theoretical Gamma Density with \alpha = 0.5');
axis([0 1 0 100]);

% The gamma_rdn function is implemented as follows:
function[w] = gamma_rdn(M,alpha)
    % Generate random numbers from the gamma distribution with parameter
    % alpha <= 1, beta = 1
    pe = exp(1);
    w = zeros(M,1);
    u = rand(100,1);
    b = (alpha + pe)/pe;
    i = 0;
    j = 0;
    while j < M
        i = i+1;
        y = b*u(i,1);
        if y <= 1
            z = y^(1/alpha);
            i = i+1;
            if u(i,1) <= exp(-z)
                j = j+1;
                w(j,1) = z;
            else
                i = i+1;
            end
        else
            z = -log((b-y)/alpha);
            i = i+1;
            if u(i,1) <= z^(alpha - 1)
                j = j+1;
                w(j,1) = z;
            else
                i = i+1;
            end
        end
    end
    if i > 95
        u = rand(100,1);
        i = 0;
    end
end

1 Answer 1

1

Is there a particular reason you chose u = rand(100,1)?

The problem is coming because in while loop, as soon as variable i exceeds 100 (say i=101), y = b*u(i,1) becomes invalid. That is, you are trying to access u(101,1) while the size of u is (100,1).

If there's not particular reason, try a large enough size, like, u = rand(10000,1).

Sign up to request clarification or add additional context in comments.

1 Comment

Thank you, this was able to fix it!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.