I learned R & MLE in the last few days. It is great! I wrote up my explorations as http://www.mayin.org/ajayshah/KB/R/mle/mle.html I will be most happy if R gurus will look at this and comment on how it can be improved. I have a few specific questions: * Should one use optim() or should one use stats4::mle()? I felt that mle() wasn't adding much value compared with optim, and in addition, I wasn't able to marry my likelihood functions to it. * One very nice feature of mle() is that you can specify a few parameters which should be fixed in the estimation. How can one persuade optim() to behave like that? * Can one use deriv() and friends to get analytical derivatives of these likelihood functions? I found I wasn't able to make headway when I was using vector/matrix notation. I think the greatness of R lies in a lovely vector/matrix notation, and it seems like a shame to have to not use that when trying to do deriv(). * For iid problems, the computation of the likelihood function and it's gradient vector are inherently parallelisable. How would one go about doing this within R? -- Ajay Shah Consultant ajayshah at mayin.org Department of Economic Affairs http://www.mayin.org/ajayshah Ministry of Finance, New Delhi
Ajay Narottam Shah wrote:>I learned R & MLE in the last few days. It is great! I wrote up my >explorations as > > http://www.mayin.org/ajayshah/KB/R/mle/mle.html > >I will be most happy if R gurus will look at this and comment on how >it can be improved. > > > >I have a few specific questions: > >* Should one use optim() or should one use stats4::mle()? > > I felt that mle() wasn't adding much value compared with optim, and > in addition, I wasn't able to marry my likelihood functions to it. > >* One very nice feature of mle() is that you can specify a few > parameters which should be fixed in the estimation. How can one > persuade optim() to behave like that? > > >give optim() a function to optimize which do not depend on those parameters ...>* Can one use deriv() and friends to get analytical derivatives of > these likelihood functions? I found I wasn't able to make headway > when I was using vector/matrix notation. I think the greatness of R > lies in a lovely vector/matrix notation, and it seems like a shame > to have to not use that when trying to do deriv(). > >* For iid problems, the computation of the likelihood function and > it's gradient vector are inherently parallelisable. How would one go > about doing this within R? > > >Kjetil -- Kjetil Halvorsen. Peace is the most effective weapon of mass construction. -- Mahdi Elmandjra -- No virus found in this outgoing message. Checked by AVG Anti-Virus.