AStA Advances in Statistical Analysis

, Volume 98, Issue 4, pp 345–369 | Cite as

Importance sampling for Kolmogorov backward equations

  • Hermann SingerEmail author
Original Paper


The solution of the Kolmogorov backward equation is expressed as a functional integral by means of the Feynman–Kac formula. The expectation value is approximated as a mean over trajectories. In order to reduce the variance of the estimate, importance sampling is utilized. From the optimal importance density, a modified drift function is derived which is used to simulate optimal trajectories from an Itô equation. The method is applied to option pricing and the simulation of transition densities and likelihoods for diffusion processes. The results are compared to known exact solutions and results obtained by numerical integration of the path integral using Euler transition kernels. The importance sampling leads to strong variance reduction, even if the unknown solution appearing in the drift is replaced by known reference solutions. In models with low-dimensional state space, the numerical integration method is more efficient, but in higher dimensions it soon becomes infeasible, whereas the Monte Carlo method still works.


Transport equation Feynman–Kac formula Monte Carlo method Drift correction Stochastic differential equations Likelihood functions Option pricing 


  1. Abrikosov, A., Gorkov, L., Dzyaloshinsky, I.: Methods of Quantum Field Theory in Statistical Physics, p. 48. Dover, New York (1963)Google Scholar
  2. Aït-Sahalia, Y.: Maximum likelihood estimation of discretely sampled diffusions: a closed-form approximation approach. Econometrica 70(1), 223–262 (2002)MathSciNetzbMATHCrossRefGoogle Scholar
  3. Aït-Sahalia, Y.: Closed-form likelihood expansions for multivariate diffusions. Ann. Stat. 36(2), 906–937 (2008)zbMATHCrossRefGoogle Scholar
  4. Beskos, A., Papaspiliopoulos, O., Roberts, G., Fearnhead, P.: Exact and efficient likelihood-based inference for discretely observed diffusion processes (with discussion). J. R. Stat. Soc. Ser. B 68, 333–382 (2006)MathSciNetzbMATHCrossRefGoogle Scholar
  5. Bibby, M., Sorensen, M.: Martingale estimation functions for discretely observed diffusion processes. Bernoulli 1, 1–39 (1995)MathSciNetCrossRefGoogle Scholar
  6. Black, F., Scholes, M.: The pricing of options and corporate liabilities. J. Polit. Econ. 81, 637–654 (1973)CrossRefGoogle Scholar
  7. Borodin, A., Salminen, P.: Handbook of Brownian Motion—Facts and Formulae, 2nd edn. Birkhäuser-Verlag, Basel (2002)Google Scholar
  8. Cameron, R.H.: A “Simpson’s Rule” for the numerical evaluation of Wiener integrals in function space. Duke Math. J. 18, 111–130 (1951)MathSciNetzbMATHCrossRefGoogle Scholar
  9. Cameron, R.H.: A family of integrals serving to connect the Wiener and Feynman integrals. J. Math. Phys. 39(126–140), 1961 (1960)Google Scholar
  10. Cameron, R.H., Martin, W.T.: Transformations of Wiener integrals under a general class of linear transformations. Trans. Am. Math. Soc. 58(2), 184–219 (1945)MathSciNetzbMATHCrossRefGoogle Scholar
  11. Chorin, A.J.: Accurate evaluation of Wiener integrals. Math. Comput. 27(121), 1–15 (1973)MathSciNetzbMATHCrossRefGoogle Scholar
  12. Cox, J., Ross, S.: The valuation of options for alternative stochastic processes. J. Financ. Econ. 3, 145–166 (1976)CrossRefGoogle Scholar
  13. Dieterich, W., Peschel, I., Schneider, W.R.: Diffusion in periodic potentials. Z. Phys. B Condens. Matter 27, 177–187 (1977)Google Scholar
  14. Durham, G.B., Gallant, A.R.: Numerical techniques for simulated maximum likelihood estimation of stochastic differential equations. J. Bus. Econ. Stat. 20, 297–316 (2002)MathSciNetCrossRefGoogle Scholar
  15. Emanuel, D., MacBeth, J.: Modelling the persistence of conditional variances. J. Financ. Quant. Anal. XVII 4, 533–554 (1982)CrossRefGoogle Scholar
  16. Feller, W.: Two singular diffusion problems. Ann. Math. 54, 173–182 (1951)MathSciNetzbMATHCrossRefGoogle Scholar
  17. Feynman, R.: Space-time approach to nonrelativistic quantum mechanics. Rev. Mod. Phys. 20, 367–387 (1948)MathSciNetCrossRefGoogle Scholar
  18. Gelfand, I., Yaglom, A.: Integration in functional spaces and its application in quantum physics. J. Math. Phys. 1(1), 48–69 (1960)CrossRefGoogle Scholar
  19. Girolami, M., Calderhead, B.: Riemann manifold Langevin and Hamiltonian Monte Carlo methods. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 73(2), 123–214 (2011)MathSciNetCrossRefGoogle Scholar
  20. Glasserman, P.: Monte Carlo Methods in Financial Engineering. Springer, New York (2003)Google Scholar
  21. Holden, H., Øksendal, B., Ubøe, J., Zhang, T.: Stochastic Partial Differential Equations: A Modeling, White Noise Functional Approach, 2nd edn, ch. 4.3, p. 146. Springer, Berlin (2009)Google Scholar
  22. Kac, M.: On distributions of certain Wiener functionals. Trans. Am. Math. Soc. 65, 1–13 (1949)zbMATHCrossRefGoogle Scholar
  23. Kac, M.: Wiener and integration in function spaces. Bull. Am. Math. Soc. 72, 52–68 (1966)zbMATHCrossRefGoogle Scholar
  24. Karatzas, I., Shreve, S.: Brownian Motion and Stochastic Calculus. Springer, Berlin (1991)Google Scholar
  25. Kloeden, P., Platen, E.: Numerical Solution of Stochastic Differential Equations. Springer, Berlin (1999) (corrected third printing)Google Scholar
  26. Kolmogoroff, A.: Über die analytischen Methoden in der Wahrscheinlichkeitsrechnung. Math. Ann. 104, 415–458 (1931)MathSciNetCrossRefGoogle Scholar
  27. Kulikov, G., Kulikova, M.: Accurate numerical implementation of the continuous–discrete extended Kalman filter. IEEE Trans. Autom. Control (2013) (in print)Google Scholar
  28. Lighthill, M.: Introduction to Fourier Analysis and Generalised Functions. Cambridge University Press, Cambridge (1958)Google Scholar
  29. Liptser, R., Shiryayev, A.: Statistics of Random Processes, vol. I, ch. 7.7, 2nd edn. Springer, Berlin (2001)Google Scholar
  30. Lorenz, E.: Deterministic nonperiodic flow. J. Atmos. Sci. 20, 130 (1963)CrossRefGoogle Scholar
  31. Nelson, E.: Feynman integrals and the Schrödinger equation. J. Math. Phys. 5(3), 332–343 (1964)Google Scholar
  32. Ozaki, T.: Non-linear time series models and dynamical systems. In: Hannan, E.J., Krishnaiah, P.R., Rao, M.M. (eds.) Handbook of Statistics, vol. 5, pp. 25–83. North Holland, Amsterdam (1985).
  33. Risken, H.: The Fokker–Planck equation, 2nd edn, ch. 4.2.1–2, p. 69. Springer, Berlin (1989)Google Scholar
  34. Singer, H.: Finanzmarktökonometrie. Zeitstetige Systeme und ihre Anwendung in Ökonometrie und empirischer Kapitalmarktforschung [Econometrics of Financial Markets, in German], ch. 9.6, pp. 248. Physica-Verlag, Heidelberg (1999)Google Scholar
  35. 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)CrossRefGoogle Scholar
  36. Singer, H.: Simulated maximum likelihood in nonlinear continuous–discrete state space models: importance sampling by approximate smoothing. Comput. Stat. 18(1), 79–106 (2003)zbMATHGoogle Scholar
  37. Singer, H.: Continuous–discrete state-space modeling of panel data with nonlinear filter algorithms. AStA Adv. Stat. Anal. 95, 375–413 (2011)MathSciNetzbMATHCrossRefGoogle Scholar
  38. Singer, H.: Maximum maximum likelihood estimation of continuous–discrete state-space models: Langevin path sampling vs. numerical integration. (2013) (preprint)Google Scholar
  39. Steinbauer, A.: Quadrature formulas for the Wiener measure. J. Complex. 15(4), 476–498 (1999)MathSciNetzbMATHCrossRefGoogle Scholar
  40. Wagner, W.: Unbiased Monte Carlo evaluation of certain functional integrals. J. Comput. Phys. 71, 21–33 (1987)MathSciNetzbMATHCrossRefGoogle Scholar
  41. Yosida, K.: Functional Analysis, pp. 419–430. Springer, Berlin (1980)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.Lehrstuhl für angewandte Statistik und Methoden der empirischen SozialforschungFernUniversität in HagenHagenGermany

Personalised recommendations