[R-SIG-Finance] (no subject)

[francesco.citta at gmail.com]

Hello,
I found a problem when I tried to use the "maxreturnPortfolio" function of fPortfolio's package (3011.78 version) after setting a target risk.
This is my code:
defaultSpec=portfolioSpec()
setOptimize(defaultSpec)="risk"
setTargetRisk(defaultSpec)=0.3
lppData=100*LPP2005.RET
defaultData=portfolioData(lppData,spec=defaultSpec)
maxReturn=maxreturnPortfolio(defaultData,spec = defaultSpec,constraints = "LongOnly")
print(maxReturn)

The output is:
Title:
 MV Return Maximized Efficient Portfolio 
 Estimator:         covEstimator 
 Solver:            solveRquadprog 
 Optimize:          maxReturn 
 Constraints:       LongOnly 
Portfolio Weights:
   SBI    SPI    SII    LMI    MPI    ALT 
0.3555 0.0000 0.0890 0.4893 0.0026 0.0636 
Covariance Risk Budgets:
   SBI    SPI    SII    LMI    MPI    ALT 
0.3555 0.0000 0.0890 0.4893 0.0026 0.0636 
Target Returns and Risks:
  mean    Cov   CVaR    VaR 
0.0105 0.0986 0.2020 0.1558 

The Target Risk (0.3) is different than optimized risk (0.0986) and the portfolio weitghts are the same of the min risk problem. For this target volatility I will expect a return near 0.05.
I tried to change the solver (setSolver(defaultSpec)="solveRsocp") but I obtained this error message: Error in as.vector(data) : no method for coercing this S4 class to a vector
Can you help me?
Thanks
Francesco Citta




	[[alternative HTML version deleted]]