+ Reply to Thread
Results 1 to 1 of 1

Thread: New guy "stumped" on a HW problem

  1. #1
    Points: 9, Level: 1
    Level completed: 17%, Points required for next Level: 41

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    New guy "stumped" on a HW problem




    Hello, I found this forum and hoped I would be able to get help with a problem that is just consuming all of my time.

    It involves finding the Maximum Likelihood Estimator. I apologize in advance for maybe not expressing any formulas correctly on here...I've never typed these in a post like this before so I don't know how to correctly type symbols for "sum" and "product" etc.

    Basically, the problem is this:

    Density function f(x) = (3/lambda) * x^2 * e^((-x^3)/lambda)

    Find the MLE of Lambda for a random sample of size n.

    So...

    I began by taking the product of the function of lambda for all values of x

    L(lambda, x1...xn) = (the product sign of i=1 to n) * f(xi, lambda) = (3/lambda) * x^2 * e^((-x^3)/lambda)

    => I take the derivative with respect to lambda...

    dln (3/lambda^n * sum of i to n for xi^2 * e^((sum of i to n of -x^3)/(lambda^n)))

    This is where I've either gone very wrong or I've just confused myself to no end.

    I think you have to take the natural log with respect to lambda and solve for lambda, but the more I look at this problem and how I'm going about it the more I confuse myself.

    If anyone could tell me how I could type this cleaner and look more like an actual problem on paper (with correct symbols, etc) I would be more than happy to go further, but this is basically where I'm stuck.

    Any help is greatly appreciated.

    Edit*

    I forgot to add the parameters

    x > 0 and lambda > 0
    Last edited by Sublime440; 10-10-2016 at 10:27 PM. Reason: forgot to add parameters

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats