[R-SIG-Finance] fPortfolio

Brian G. Peterson brian at braverock.com
Thu Dec 11 03:30:07 CET 2008


There is not always a solution if you set your constraints too high.  If the riskfree rate is set to zero, do you get a rational result? If you use different assets, does the portfdolio behave? Etc.

If you post your instrument data or historical data download code, I'm sure saomone will try to replicate your error, but with only the code and not the data, we're just guessing (see list posting guidelines)

Regards,

     - Brian

Bastian Offermann <bastian2507hk at yahoo.co.uk> wrote:

>Nobody?? I am wondering what the problem is. Using different inputs for 
>setRiskFreeRate(Spec)  I always get the same output (i.e. same weights). 
>And there is never an allocation to the riskfree asset no matter what my 
>required return is? Where is my mistake? Am I totally wrong here??
>
>Thanks.
>
>Regards
>
>B
>
>
>Hello,
>
>i have encountered a problem with fPortfolio recently. I am using below 
>code:
>
>the problem is that my risk free rate is different from the one 
>displayed in the graphical output and the straight line in the graphical 
>output intersects the efficient frontier twice. same problem when i use 
>"cmlLines(eff.front)" .what am i missing out on?
>
>THanks.
>
>Data1 <- as.timeSeries(Data1)
>Data1 <- as.timeSeries(data(LPP2005REC))
>Data1 <- Data1[,1:4]*250
>
>r.p <- 0.10
>r.rf <- 0.01
>
>Spec <- portfolioSpec()
>restr <- c("Short", "LongOnly")
>cons <- restr[1]
>setOptimize(Spec) <- "minRisk"
>setRiskFreeRate(Spec) <- r.rf #/250
>setTargetReturn(Spec) <- r.p #/250
>setNFrontierPoints(Spec) <- 50
>solver <- c("solveRquadprog", "solveRshortExact")
>if(cons == "LongOnly") { setSolver(Spec) <- solver[1] }
>if(cons == "Short") { setSolver(Spec) <- solver[2] }
>
>eff.front <- portfolioFrontier(data = Data1, spec = Spec, constraints = 
>cons)
>eff.port <- efficientPortfolio(data = Data1, spec = Spec, constraints = 
>cons)
>tang.port <- tangencyPortfolio(data = Data1, spec = Spec, constraints = 
>cons)
>minvar.port <- minvariancePortfolio(data = Data1, spec = Spec, 
>constraints = cons)
>w <- as.vector(getWeights(eff.port))
>w <- matrix(w, 1, ncol(Data1))
>w.min <- as.vector(getWeights(minvar.port))
>w.min <- matrix(w.min, 1, ncol(Data1))
>w
>w.min
>frontierPlot(eff.front, frontier = "upper", return="mu", risk="Sigma", 
>type="l")
>minvariancePoints(eff.front, col="red", bg="red", pch=21)
>tangencyLines(eff.front)
>tangencyPoints(eff.front)
>
>_______________________________________________
>R-SIG-Finance at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>-- Subscriber-posting only.
>-- If you want to post, subscribe first.


More information about the R-SIG-Finance mailing list