Applying Genetic Programming in Business Forecasting

Created by W.Langdon from gp-bibliography.bib Revision:1.4215

  author =       "Arthur K. Kordon",
  title =        "Applying Genetic Programming in Business Forecasting",
  booktitle =    "Genetic Programming Theory and Practice XI",
  year =         "2013",
  series =       "Genetic and Evolutionary Computation",
  editor =       "Rick Riolo and Jason H. Moore and Mark Kotanchek",
  publisher =    "Springer",
  chapter =      "6",
  pages =        "101--117",
  address =      "Ann Arbor, USA",
  month =        "9-11 " # may,
  keywords =     "genetic algorithms, genetic programming, Nonlinear
                 forecasting, Symbolic regression, Nonlinear transforms,
                 Business applications",
  isbn13 =       "978-1-4939-0374-0",
  DOI =          "doi:10.1007/978-1-4939-0375-7_6",
  abstract =     "Since the global recession of 2008-2009, it has been
                 much more widely understood that reliable economic
                 forecasting is needed in business decision-making. Of
                 special interest are the forecasting methods based on
                 explanatory variables (economic drivers), the most
                 popular of which is the Auto-Regressive Integrated
                 Moving-Average with eXplanatory variables (ARIMAX)
                 model. A limitation of this approach, however, is the
                 assumption of linear relationships between the
                 explanatory variables and the target variable. Genetic
                 programming is a potential solution for representing
                 nonlinearity and a hybrid scheme of integrating static
                 and dynamic nonlinear transforms into the ARIMAX models
                 is proposed in the chapter. From an implementation
                 point of the view the proposed solution has several
                 advantages, such as: optimal synergy between two
                 well-known approaches like GP and ARIMAX, avoiding the
                 need for developing a solid theoretical alternative for
                 nonlinear time series modelling, using available
                 forecasting software, and low efforts to train the
                 final user. The proposed approach is illustrated with
                 two examples from real business applications in the
                 area of raw materials forecasting.",
  notes =        "

                 Part of \cite{Riolo:2013:GPTP} published after the
                 workshop in 2013",

Genetic Programming entries for Arthur K Kordon