> Ranjan Maitra
> on Sun, 27 Aug 2017 08:17:15 -0500 writes:
> I have not followed the history of this thread, but I am
> quite flummoxed as to why the OP is rewriting code to
> estimate parameters from an univariate Gaussian mixture
> model when alternatives such as E
Just a quick note:
"I have tried to use optimr but it gives me the local maxima, now I am
struck with the problem of how to get the global maxima"
This is a profound misunderstanding of how numerical optimization
works. **All** numerical optimizers can provide only "local" maxima
(the exact meani
I have not followed the history of this thread, but I am quite flummoxed as to
why the OP is rewriting code to estimate parameters from an univariate Gaussian
mixture model when alternatives such as EMCluster (which generally appears to
handle initialization better than MClust) exist. Or perhaps
Dear Niharika,
As I said before, the problem is basically an optimization issue. You should
isolate the problematic part from the rest of your study. Sometimes, more
information does not help to solution. All the answers from us (Ulrik, David,
me) are more or less are correct to find a maximum
4 matches
Mail list logo
