+ Reply to Thread
Results 1 to 4 of 4

Thread: Estimation Maximization

  1. #1
    Points: 43, Level: 1
    Level completed: 86%, Points required for next Level: 7

    Posts
    5
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Estimation Maximization




    Hello everybody!

    I have a general question concerning Estimation Maximization (EM): I understand from literature research, that EM is not possible for dichotomous variables (like "high" vs. "low" for example). Is that assumption right?

    Thank you for any help!
    therp

  2. #2
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Location
    Konstanz
    Posts
    372
    Thanks
    3
    Thanked 146 Times in 123 Posts

    Re: Estimation Maximization

    I assume you mean expectation maximization. In that case your assumption is incorrect. It is however an extremely general technique, and it may be that in the particular software implementation that you use a dichotomous variable would be a problem.

  3. #3
    Points: 43, Level: 1
    Level completed: 86%, Points required for next Level: 7

    Posts
    5
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: Estimation Maximization

    I'm sorry!! Yes, that is what I meant .
    Exactly, SPSS doesn't produce estimated values when running EM for dichotomous variables. Do you have any advice on other software that does that? How is the mechanism different in EM for dichotomous variables?
    I could impute missing dichotomous values by treating dichotomous variables as continuous in SPSS, but I guess that's "cheating"...any comments on that?
    thank you!!

  4. #4
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Location
    Konstanz
    Posts
    372
    Thanks
    3
    Thanked 146 Times in 123 Posts

    Re: Estimation Maximization


    That depends on what you want to do. The EM algorithm is a very general for finding the maximum likelihood or maximum a-posteriori estimates of some model. In order for us to be able to help you, we need to know what model you whish to estimate.

+ 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