Rebalancing triggers for the CPPI using genetic programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4208

  author =       "Tikesh Ramtohul and Dietmar Maringer",
  title =        "Rebalancing triggers for the CPPI using genetic
  booktitle =    "Proceedings of the 4th International Conference on
                 Computational and Financial Econometrics CFE'10",
  year =         "2010",
  editor =       "G. Kapetanios and O. Linton and M. McAleer and 
                 E. Ruiz",
  pages =        "E515",
  address =      "Senate House, University of London, UK",
  month =        "10-12 " # dec,
  organisation = "CSDA, LSE, Queen Mary and Westerfield College",
  publisher =    "ERCIM",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "",
  abstract =     "The Constant Proportion Portfolio Insurance (CPPI)
                 technique is a dynamic capital-protection strategy that
                 aims at providing investors with a guaranteed minimum
                 level of wealth at a specified time horizon. It gives
                 an investor the ability to limit downside risk while
                 allowing some participation in upside markets. Despite
                 its widespread popularity in the industry, this
                 strategy is not bereft of risk for the issuer. The
                 inability to meet the guarantee at maturity, commonly
                 referred to as gap risk, is a major concern. Gap risk
                 could be limited by increasing the frequency of trading
                 (rebalancing) but this has the adverse effect of
                 generating more transaction costs. One way of achieving
                 a tradeoff between gap risk and rebalancing frequency
                 is through the use of rebalancing triggers, which
                 constitutes the main focus of this paper. We use a
                 genetic programming (GP) approach to obtain bounds of
                 tolerance around the value of the multiplier implied by
                 the portfolio composition. The experiments focus on GBM
                 and GARCH price processes, and two different types of
                 fitness functions, namely Tracking Error and Sortino
                 ratio. We investigate the performance of the GP-based
                 rebalancing strategy for different parameter settings
                 and find that it yields better results than
                 calendar-based rebalancing strategies in most
  notes =        "Abstracts only

                 WWZ Uni Basel, Switzerland

Genetic Programming entries for Tikesh Ramtohul Dietmar G Maringer