1

I am researching a software reliability model that has many parameters that need to be estimated. I want to use a maximum likelihood estimator (MLE) to estimate those parameters, but my model is too complicated to derive the log-likelihood function (namely LL in the following picture) from the likelihood function (namely L in the following picture).

When I try to execute Log[L] in Mathematica, I don't get the expected answer (LL). So my question is, how can I get expected answer after taking the log of the equation? Or can Mathematica not do this operation?

Link of code:https://www.dropbox.com/s/e81ndt3kgdq5dsw/GO_model.nb

enter image description here

Steven Chen
  • 53
  • 4
  • 3
    Please look up Simplify, FullSimplify and ComplexityFunction. Also search this site for useful posts on those. – rm -rf Sep 16 '13 at 16:37
  • 2
    Can you add the exact code instead the pictures of it ? – Sektor Sep 16 '13 at 16:59
  • Sorry, I am rookie in here, so I directly post the link for code in the picture. – Steven Chen Sep 16 '13 at 17:16
  • I try to input SuperLog[On],but I just get output SuperLog[On]. – Steven Chen Sep 16 '13 at 17:34
  • @user2336074 Unfortunately the SuperLog thing referenced in the top answer is only available by purchasing the add-on package mathStatica. – Mr.Wizard Sep 16 '13 at 17:41
  • @Mr.Wizard♦ I find the paper http://epub.ub.uni-muenchen.de/1530/1/paper_141.pdf, and it said "This package is also available electronically by anonymous ftp at ftp://ftp.tri.org.au/SMLE.m". But,I still can't download the package from ftp://ftp.tri.org.au/SMLE.m. – Steven Chen Sep 16 '13 at 18:00
  • Just use the replacement rule in the question, it works fine for your expression. – ssch Sep 16 '13 at 18:01
  • @ssch Sorry I don't understand what replacement rule means. And can I use SuperLog for free? Thanks for your help. – Steven Chen Sep 16 '13 at 18:23

0 Answers0