• DocumentCode
    3256990
  • Title

    Piecewise constant modeling and Kalman filter tracking of systematic market risk

  • Author

    Rajbhandary, Triloke ; Xiao-Ping Zhang ; Fang Wang

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Ryerson Univ., Toronto, ON, Canada
  • fYear
    2013
  • fDate
    3-5 Dec. 2013
  • Firstpage
    1144
  • Lastpage
    1144
  • Abstract
    In this paper, we present a new piecewise constant model to represent time-varying systematic risk, i.e., beta. We develop a new tracking algorithm for the new model based on modified Kalman filter that uses Bayes´ criteria. Empirical results show the superiority of our method over traditional random walk, mean reverting and moving window beta estimates.
  • Keywords
    Bayes methods; Kalman filters; marketing; risk analysis; tracking filters; Bayes criteria; mean reverting estimates; modified Kalman filter tracking algorithm; moving window beta estimates; piecewise constant modeling; random walk; time-varying systematic market risk; Adaptation models; Computational modeling; Economics; Educational institutions; Equations; Kalman filters; Systematics;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Global Conference on Signal and Information Processing (GlobalSIP), 2013 IEEE
  • Conference_Location
    Austin, TX
  • Type

    conf

  • DOI
    10.1109/GlobalSIP.2013.6737107
  • Filename
    6737107