Skip to main content
SpringerLink
Log in

Generalized Gauss–Hermite filtering

  • Original Paper
  • Published:
AStA Advances in Statistical Analysis Aims and scope Submit manuscript

Abstract

We consider a generalization of the Gauss–Hermite filter (GHF), where the filter density is represented by a Hermite expansion with leading Gaussian term (GGHF). Thus, the usual GHF is included as a special case. The moment equations for the time update are solved stepwise by Gauss–Hermite integration, and the measurement update is computed by the Bayes formula, again using numerical integration. The performance of the filter is compared numerically with the GHF, the UKF (unscented Kalman filter) and the EKF (extended Kalman filter) and leads to a lower mean squared filter error.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Abramowitz, M., Stegun, I.: Handbook of Mathematical Functions. Dover, New York (1965)

    Google Scholar 

  • Aït-Sahalia, Y.: Maximum likelihood estimation of discretely sampled diffusions: a closed-form approximation approach. Econometrica 70(1), 223–262 (2002)

    Article  MathSciNet  MATH  Google Scholar 

  • Arnold, L.: Stochastic Differential Equations. Wiley, New York (1974)

    MATH  Google Scholar 

  • Challa, S., Bar-Shalom, Y., Krishnamurthy, V.: Nonlinear filtering via generalized edgeworth series and Gauss-Hermite quadrature. IEEE Trans. Signal Process. 48, 1816–1820 (2000)

    Article  Google Scholar 

  • Courant, R., Hilbert, D.: Methoden der Mathematischen Physik. Springer, Berlin (1968)

    Google Scholar 

  • Daum, F.: Exact finite-dimensional nonlinear filters. IEEE Trans. Autom. Control 31, 616–622 (1986)

    Article  MathSciNet  MATH  Google Scholar 

  • Daum, F.: Nonlinear filters: beyond the Kalman filter. IEEE A&E Syst. Mag 20(8), 57–69 (2005)

    Article  Google Scholar 

  • Ito, K., Xiong, K.: Gaussian filters for nonlinear filtering problems. IEEE Trans. Autom. Control 45(5), 910–927 (2000)

    Article  MathSciNet  MATH  Google Scholar 

  • Jazwinski, A.: Stochastic Processes and Filtering Theory. Academic, New York (1970)

    MATH  Google Scholar 

  • Julier, S., Uhlmann, J.: A new extension of the Kalman filter to nonlinear systems. In: The 11th International Symposium on Aerospace/Defense Sensing, Simulation and Control, Orlando, Florida (1997)

  • Julier, S., Uhlmann, J.: Unscented Kalman filtering and nonlinear estimation. Proc. IEEE 92(3), 401–422 (2004)

    Article  Google Scholar 

  • Julier, S., Uhlmann, J., Durrant-White, H.F.: A new method for the nonlinear transformation of means and covariances in filters and estimators. IEEE Trans. Autom. Control 45(3), 477–482 (2000)

    Article  MATH  Google Scholar 

  • Kuznetsov, P., Stratonovich, R., Tikhonov, V.: Quasi-moment functions in the theory of random processes. Theory Probab. Appl. V(1), 80–97 (1960)

    Article  Google Scholar 

  • Liptser, R., Shiryayev, A.: Statistics of Random Processes, vols. I, II, 2nd edn. Springer, New York (2001)

    Google Scholar 

  • Miller, R., Ghil, M., Gauthiez, F.: Advanced data assimilation in strongly nonlinear dynamical systems. J. Atmos. Sci. 51(8), 1037–1056 (1994)

    Article  MathSciNet  Google Scholar 

  • Nørgaard, M., Poulsen, N., Ravn, O.: New developments in state estimation for nonlinear systems. Automatica 36, 1627–1638 (2000)

    Article  Google Scholar 

  • Singer, H.: Parameter estimation of nonlinear stochastic differential equations: simulated maximum likelihood vs. extended Kalman filter and Itô-Taylor expansion. J. Comput. Graph. Stat. 11(4), 972–995 (2002)

    Article  Google Scholar 

  • Singer, H.: Continuous-discrete unscented Kalman filtering. Diskussionsbeiträge Fachbereich Wirtschaftswissenschaft 384, FernUniversität in Hagen. http://www.fernuni-hagen.de/FBWIWI/forschung/beitraege/pdf/db384.pdf (2006a)

  • Singer, H.: Generalized Gauss-Hermite filtering for multivariate diffusion processes. Diskussionsbeiträge Fachbereich Wirtschaftswissenschaft 402, FernUniversität in Hagen. http://www.fernuni-hagen.de/FBWIWI/forschung/beitraege/pdf/db402.pdf (2006b)

  • Singer, H.: Moment equations and Hermite expansion for nonlinear stochastic differential equations with application to stock price models. Comput. Stat. 21(3), 385–397 (2006c)

    Article  MATH  Google Scholar 

  • Singer, H.: Stochastic differential equation models with sampled data. In: van Montfort, K., Oud, H., Satorra, A. (eds.) Longitudinal Models in the Behavioral and Related Sciences. The European Association of Methodology (EAM) Methodology and Statistics Series, vol. II, pp. 73–106. Lawrence Erlbaum Associates, Mahwah (2007)

    Google Scholar 

  • Srinivasan, K.: State estimation by orthogonal expansion of probability distributions. IEEE Trans. Autom. Control 15(1), 3–10 (1970)

    Article  Google Scholar 

  • Tam, W., Hatzinakos, D.: An efficient radar tracking algorithm using multidimensional Gauss-Hermite quadratures. IEEE Int. Conf. Acoust. Speech Signal Process. 5, 3777–3780 (1997)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Lehrstuhl für Angewandte Statistik und Methoden der Empirischen Sozialforschung, FernUniversität in Hagen, 58084, Hagen, Germany

    Hermann Singer

Authors
  1. Hermann Singer
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Hermann Singer.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Singer, H. Generalized Gauss–Hermite filtering. AStA 92, 179–195 (2008). https://doi.org/10.1007/s10182-008-0068-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10182-008-0068-z

Keywords

Search

Navigation