RollingPortfolio         package:fPortfolio         R Documentation

_R_o_l_l_i_n_g _P_o_r_t_f_o_l_i_o

_D_e_s_c_r_i_p_t_i_o_n:

     A collection and description of functions  allowing to roll a
     portfolio optimization over time. 

     The functions are:

       'rollingWindows'               Returns a list of rolling window frames,
       'rollingCmlPortfolio'          Rolls a CML portfolio,
       'rollingTangencyPortfolio'     Rolls a tangency portfolio,
       'rollingMinvariancePortfolio'  Rolls a minimum risk portfolio,
       'rollingPortfolioFrontier'     returns an efficient portfolio,
       'portfolioBacktesting'         Does portfolio backtesting.

_U_s_a_g_e:

     rollingWindows(x, period = "12m", by = "1m")

     rollingCmlPortfolio(data, spec, constraints, from, to, action = NULL, 
         title = NULL, description = NULL, ...)
     rollingTangencyPortfolio(data, spec, constraints, from, to, action = NULL, 
         title = NULL, description = NULL, ...)
     rollingMinvariancePortfolio(data, spec, constraints, from, to, action = NULL, 
         title = NULL, description = NULL, ...)
         
     rollingPortfolioFrontier(data, spec, constraints, from, to, action = NULL, 
         title = NULL, description = NULL, ...)
        
     portfolioBacktesting(formula, data, spec = portfolioSpec(), constraints = NULL, 
         portfolio = "minvariancePortfolio", horizon = "12m", smoothing = "6m", 
         trace = TRUE)

_A_r_g_u_m_e_n_t_s:

  action: [*Portfolio][*Frontier] - 
           a  character string naming a user defined function. This
          function is optionally applied after each rolling step. 

      by: [rollingWindows] - 
           a character string, by default '"1m"', which denotes 1
          month. The shift by which the portfolio is rolled. 

constraints: [*Portfolio][*Frontier] - 
           a character string vector, containing the constraints of the
          form
           '"minW[asset]=percentage"' for box constraints resp. 
           '"maxsumW[assets]=percentage"' for sector constraints. 

    data: [*Portfolio][*Frontier] - 
           a list, having a statistics named list, having named entries
          'mu' and  'Sigma', containing the information of the
          statistics. 
           [*Backtesting] - 


description: [*Portfolio][*Frontier] - 
           a character string, allowing for a brief project
          description, by default NULL, i.e. Date and User. 

 formula: [*Backtesting] - 
           a formula describing the benchmark and assets used for
          backtesting in the form 'backtest ~ assetA + ... + assetZ'.
          Here, 'backtest' and 'asset*' are column names of the 'data'
          set. 

from, to: [*Portfolio][*Frontier] - 
           a vector of S4 'timeDate' objects which denote the starting
          and ending dates for the investigation. 

 horizon: [*Backtesting] - 
           a character string, by default '"12m"', which denotes 12
          months. The period over which the portfolio is rolled. 

  period: [rollingWindows] - 
           a character string, by default '"12m"', which denotes 12
          months. The period over which the portfolio is rolled. 

portfolio: [portfolioBacktesting] - 
           a character string, the function name of a user defined 
          portfolio function, which defines the rolling portfolio
          strategy. 

smoothing: [portfolioBacktesting] - 
           a character string, by default '"6m"', which denotes 6
          months. The period over which the rolling weights are
          smoothed by an exponential moving average. 

    spec: [*Portfolio][*Frontier] - 
           an S4 object of class 'fPFOLIOSPEC', containing slots call, 
          model, portfolio, title, description,  see 'PortfolioSpec'
          for a full slot description. 

   title: [*Portfolio][*Frontier] - 
           a character string, containing the title for the object, by 
          default NULL. 

   trace: [portfolioBacktesting][*OptimalPortfolio] - 
           a logical value. Should the calculataion be traced? 

       x: [rollingWindows] - 
           an S4 object of class 'timeSeries' from which the rolling
          window frames will be created. The length of these frames is
          given by the argument 'period' and they are shifted by the
          value specified by the argument 'by'. 

     ...: optional arguments to be passed.


_D_e_t_a_i_l_s:

     *RollingWindows:* The function 'rollingWindows' constructs from a
     'timeSeries' object windows frames of given length 'period' and
     shift 'by'. ... 

     *Rolling Portfolios:* 

      The functions 'rolling*Portfolio' ... 

     *Rolling Frontier:* 

         The function 'rollingPortfolioFrontier' ...

     *Rolling Backtesting:* 

         The functions  ...

_V_a_l_u_e:

     'rollingwindows()'
      returns ... 

     'rollingCmlPortfolio'
      'rollingTangencyPortfolio'
      'rollingMinvariancePortfolio'
      return ... 

     'rollingPortfolioFrontier'
      returns ... 

     'portfolioBacktesting'
      returns ...

_A_u_t_h_o_r(_s):

     Diethelm Wuertz and Oliver Greshake for the Rmetrics port.

_S_e_e _A_l_s_o:

     'PortfolioClass', 'PortfolioData', 'PortfolioSpec',
     'PortfolioConstraints'.

_E_x_a_m_p_l_e_s:

     ## ...

