Hi - First posting here.
I am using fPortfolio to try and optimize a simple portfolio consisting of 5
daily return series. I want to maximize return subject to
setTargetRisk(myspec)=0.08 using only constraints="LongOnly"
I can run feasiblePortfolio() using a spec file that specifies the weights, and
it works fine.
When I run maxreturnPortfolio(mydata,myspec,"LongOnly"), however, I get
Error in .rquadprog(Dmat = args$Dmat, dvec = args$dvec, Amat = args$Amat, :
NA/NaN/Inf in foreign function call (arg 8)
Troubleshooting, it appears to be occurring in the routine eqsumWConstraints
within .rquadprogArguments. Specifically, when that routine calls
ceq <- c(Return = targetReturn, Budget = -1)
it produces
> ceq
Return Budget
NA -1
And that NA causes .rquadprog to spew, I think, when it reaches the line
optim = .Fortran("qpgen2", as.double(Dmat), dvec = as.double(dvec),
as.integer(n), as.integer(n), sol = as.double(rep(0,
n)), crval = as.double(0), as.double(Amat), as.double(bvec),
as.integer(n), as.integer(q), as.integer(meq), iact = as.integer(rep(0,
q)), nact = as.integer(0), iter = as.integer(rep(0,
2)), work = as.double(work), ierr = as.integer(0),
PACKAGE = "quadprog")
...where I have highlighted the offending argument (8).
Now, there is nothing unusual that I can see about my data. Unfortunately with
the .Fortran I don't think I can troubleshoot any further. I am hoping someone
here has a clue...
Thanks in advance.
Mike Ashton
________________________________
This email and any attachments are confidential and inte...{{dropped:10}}
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
