Anhai Jin
2013-Jun-11  19:14 UTC
[R] Package for maximizing likelihood function with EM algorithm
Hi R users, I am trying to figure out if there is a package in R that can maximize likelihood function with EM algorithm. Right now, I have derived the log-likelihood function, which is a function of 9 indicator variables with 14 parameters. Is there a package that I can specify the log-likelihood function, along with all variables and parameters, and will generate estimates? Thanks for help! [[alternative HTML version deleted]]
Berend Hasselman
2013-Jun-11  20:14 UTC
[R] Package for maximizing likelihood function with EM algorithm
On 11-06-2013, at 21:14, Anhai Jin <ajin at iknowtion.com> wrote:> Hi R users, > > I am trying to figure out if there is a package in R that can maximize likelihood function with EM algorithm. Right now, I have derived the log-likelihood function, which is a function of 9 indicator variables with 14 parameters. Is there a package that I can specify the log-likelihood function, along with all variables and parameters, and will generate estimates? >Have a look at package SQUAREM Berend
Possibly Parallel Threads
- maximizing a likelihood function containing an integral
- maximizing the gamma likelihood
- simultaneously maximizing two independent log likelihood functions using mle2
- EM algorithm to fit circular mix of uniform+Von Mises
- DOING CLUSTERING WITH THE EM ALGORITHM