[R] Constraint optimization containing constraints with absolute values

[Dimitris.Kapetanakis]

Dear all,

I would like to use constraint optimization routine but with the constraint
containing absolute values of the arguments. 

For example I would like to maximize a log-likelihood function (loglik) wrt
b where b is a vector of length=2 with the constraint abs(b[1])+abs(b[2])=2. 

The unconstraint optimization could be given by:

loglik	<- function(b) -colMeans(Y*log(Prob1(b))+(1-Y)*log(1-Prob1(b)))
ML		<- maxLik(loglik, start=matrix(1,k,1), method="NM")

but I do not know which package or code can provide me with the solution of
this particular optimization problem.

Thank you all

Dimitris


--
View this message in context: http://r.789695.n4.nabble.com/Constraint-optimization-containing-constraints-with-absolute-values-tp4502083p4502083.html
Sent from the R help mailing list archive at Nabble.com.