Question for the Tutors: Maximum Likelihood estimation vs ordinary least squares

Discussion in 'CT3' started by mugono, Sep 17, 2008.

  1. mugono

    mugono Ton up Member

    Hi,

    I was wondering whether somebody could answer a question I have to do with finding optimal estimates.

    I notice that the method of maximum likelihood appears to be the method of choice when trying to obtain optimal estimates. However, this is only possible if we make an assumption about the distribution (which has implications if our assumptions are wrong!). I would have thought for this reason that the method of least squares would always be 'preferable' as no assumption about the distribution is necessary.

    Would it be possible for somebody to shed some light on this for me?

    Thank you in advance.
     
  2. Neetu Verma

    Neetu Verma Member

    hi,
    maximum likelihood & least square method gives similar result if and only if sample taken form the normal distribution ,not in general case .
     
  3. jatin.agrawal

    jatin.agrawal Member

    will you please tell me what is least sqaure method...is it in our CT3 material..if yes then please let me know in which chapter...

    Thanks and Regards
    Jatin Agrawal
     

Share This Page