[R] finding the MLEs of IG parameters by EM-Alorithm

[peter dalgaard]

On Aug 2, 2012, at 16:01 , Sara S wrote:

> Dear all
> 
> I'm trying to caculate the MLEs for parameters of Inverse Gaussian
> distribution (in a k-sample problem with common mean) by using
> EM-Algorithm.  I found some package for EM-Algorithm that are useful
> for missing or incomplete data and are not helpful for solving my
> problem.
> (Exactly, the problem is: Let Xij, i=1,..,k , j=1,...,ni, be a random
> sample from IG(μ,λi). So the log-likelihood function is:
> log(L)=∑(ni/2)*log(λi)-∑∑ (λi/2*(μ^2)*xij)*((xij-μ)^2)+const.
> Find the MLEs for (μ,λ1,...,λk) by EM-Algorithm?)
> 
> Could you please help me with that?

No. Its a theory problem, not an R one. You likely need to rephrase the problem as a missing-data problem. I don't see how, but the text containing the exercise may contain a hint somewhere in the vicinity.

> I would like to thank you in advance for your help.
> 
> best regards,
> Sara

-- 
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com