Langevin diffusions on the torus: estimation and applications

Statistics and Computing volume 29, pages1–22(2019)Cite this article

Abstract

We introduce stochastic models for continuous-time evolution of angles and develop their estimation. We focus on studying Langevin diffusions with stationary distributions equal to well-known distributions from directional statistics, since such diffusions can be regarded as toroidal analogues of the Ornstein–Uhlenbeck process. Their likelihood function is a product of transition densities with no analytical expression, but that can be calculated by solving the Fokker–Planck equation numerically through adequate schemes. We propose three approximate likelihoods that are computationally tractable: (i) a likelihood based on the stationary distribution; (ii) toroidal adaptations of the Euler and Shoji–Ozaki pseudo-likelihoods; (iii) a likelihood based on a specific approximation to the transition density of the wrapped normal process. A simulation study compares, in dimensions one and two, the approximate transition densities to the exact ones, and investigates the empirical performance of the approximate likelihoods. Finally, two diffusions are used to model the evolution of the backbone angles of the protein G (PDB identifier 1GB1) during a molecular dynamics simulation. The software package sdetorus implements the estimation methods and applications presented in the paper.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Notes

  1. 1.

    Note that \(-(x_2-x_1\text {e}^{-(t_2-t_1)}+2k\pi \text {e}^{-t_2})^2\) should be in the exponential’s denominator of Liu (2013)’s (15) and (16).

  2. 2.

    Note the similar argument given in Roberts and Stramer (2002), albeit in their equation (24) the covariance matrix is not symmetric, probably because of a typo in (25), which should have been \((J(x)a_{x,h})'=J(x)a_{x,h}\).

  3. 3.

    In Shoji and Ozaki (1998) the drift approximation is done by Itô’s formula. To obtain a simpler pseudo-likelihood, we use a local linear approximation of b as in Ozaki (1985) (for the case \(p=1\)). Without this extra simplification, the expectation becomes \(\tilde{E}_\varDelta ({\varvec{\varphi }})=E_\varDelta ({\varvec{\varphi }})+J({\varvec{\varphi }})^{-2}(\exp \{J({\varvec{\varphi }})\varDelta \}-{\mathbf {I}}-J({\varvec{\varphi }})\varDelta )M({\varvec{\varphi }})\) with \(M({\varvec{\varphi }})=\frac{1}{2}\left( \mathrm {tr}\left[ {\mathbf {V}}({\varvec{\varphi }}){\mathbf {H}}_1({\varvec{\varphi }})\right] ,\ldots ,\mathrm {tr}\left[ {\mathbf {V}}({\varvec{\varphi }}){\mathbf {H}}_n({\varvec{\varphi }})\right] \right) '\) and \({\mathbf {H}}_i({\varvec{\varphi }})=\left( \tfrac{\partial ^2b_i({\varvec{\varphi }})}{\partial \phi _k\partial \phi _l}\right) _{1\le k,l\le p}\), \(i=1,\ldots ,p\).

References

  1. Banerjee, A., Dhillon, I.S., Ghosh, J., Sra, S.: Clustering on the unit hypersphere using von Mises–Fisher distributions. J. Mach. Learn. Res. 6, 1345–1382 (2005)

    MathSciNet  MATH  Google Scholar 

  2. Bernstein, D.S., So, W.: Some explicit formulas for the matrix exponential. IEEE Trans. Autom. Control 38(8), 1228–1232 (1993)

    MathSciNet  Article  MATH  Google Scholar 

  3. Beskos, A., Papaspiliopoulos, O., Roberts, G.O., Fearnhead, P.: Exact and computationally efficient likelihood-based estimation for discretely observed diffusion processes. J. R. Stat. Soc. Ser. B Stat. Methodol. 68(3), 333–382 (2006)

    MathSciNet  Article  MATH  Google Scholar 

  4. Beskos, A., Papaspiliopoulos, O., Roberts, G.O.: Retrospective exact simulation of diffusion sample paths with applications. Bernoulli 12(6), 1077–1098 (2006)

    MathSciNet  Article  MATH  Google Scholar 

  5. Bibby, B.M., Sørensen, M.: Simplified estimating functions for diffusion models with a high-dimensional parameter. Scand. J. Stat. 28(1), 99–112 (2001)

    MathSciNet  Article  MATH  Google Scholar 

  6. Bladt, M., Finch, S., Sørensen, M.: Simulation of multivariate diffusion bridges. J. R. Stat. Soc.: Ser. B Stat. Methodol. 78(2), 343–369 (2016)

    MathSciNet  Article  Google Scholar 

  7. Bottaro, S., Lindorff-Larsen, K., Best, R.B.: Variational optimization of an all-atom implicit solvent force field to match explicit solvent simulation data. J. Chem. Theory Comput. 9(12), 5641–5652 (2013)

    Article  Google Scholar 

  8. Breckling, J.: The Analysis of Directional Time Series: Applications to Wind Speed and Direction. Lecture Notes in Statistics, vol. 61. Springer, Berlin (1989)

  9. Codling, E., Hill, N.: Calculating spatial statistics for velocity jump processes with experimentally observed reorientation parameters. J. Math. Biol. 51(5), 527–556 (2005)

    MathSciNet  Article  MATH  Google Scholar 

  10. Dacunha-Castelle, D., Florens-Zmirou, D.: Estimation of the coefficients of a diffusion from discrete observations. Stochastics 19(4), 263–284 (1986)

    MathSciNet  Article  MATH  Google Scholar 

  11. Dehay, D.: Parameter maximum likelihood estimation problem for time periodic modulated drift Ornstein Uhlenbeck processes. Stat. Inference Stoch. Process. 18(1), 69–98 (2015)

    MathSciNet  Article  MATH  Google Scholar 

  12. Dehling, H., Franke, B., Kott, T.: Drift estimation for a periodic mean reversion process. Stat. Inference Stoch. Process. 13(3), 175–192 (2010)

    MathSciNet  Article  MATH  Google Scholar 

  13. Émery, M.: Stochastic Calculus in Manifolds. Universitext, Springer, Berlin (1989)

    Google Scholar 

  14. Frank, T.D.: Nonlinear Fokker–Planck Equations: Fundamentals and Applications. Springer Series in Synergetics. Springer, Berlin (2005)

    Google Scholar 

  15. Hill, N., Häder, D.P.: A biased random walk model for the trajectories of swimming micro-organisms. J. Theor. Biol. 186(4), 503–526 (1997)

    Article  Google Scholar 

  16. Hsu, E.P.: Stochastic analysis on manifolds, Graduate Studies in Mathematics, vol. 38. American Mathematical Society, Providence (2002)

    Google Scholar 

  17. Iacus, S.M.: Simulation and Inference for Stochastic Differential Equations: With R Examples. Springer Series in Statistics. Springer, New York (2008)

  18. In ’t Hout KJ, Foulon S.: ADI finite difference schemes for option pricing in the Heston model with correlation. Int. J. Numer. Anal. Model. 7(2), 303–320 (2010)

  19. Jammalamadaka, S.R., SenGupta, A.: Topics in Circular Statistics, Series on Multivariate Analysis, vol. 5. World Scientific Publishing, River Edge (2001)

    Google Scholar 

  20. Jona-Lasinio, G., Gelfand, A., Jona-Lasinio, M.: Spatial analysis of wave direction data using wrapped Gaussian processes. Ann. Appl. Stat. 6(4), 1478–1498 (2012)

    MathSciNet  Article  MATH  Google Scholar 

  21. Jones, M.C., Pewsey, A.: A family of symmetric distributions on the circle. J. Am. Stat. Assoc. 100(472), 1422–1428 (2005)

    MathSciNet  Article  MATH  Google Scholar 

  22. Kato, S.: A Markov process for circular data. J. R. Stat. Soc.: Ser. B Stat. Methodol. 72(5), 655–672 (2010)

    MathSciNet  Article  Google Scholar 

  23. Kent, J.: Discussion of paper by K. V. Mardia. J. R. Stat. Soc. Ser. B 37(3), 377–378 (1975)

    Google Scholar 

  24. Kent, J.: Time-reversible diffusions. Adv. Appl. Probab. 10(4), 819–835 (1978)

    MathSciNet  Article  MATH  Google Scholar 

  25. Kessler, M.: Simple and explicit estimating functions for a discretely observed diffusion process. Scand. J. Stat. 27(1), 65–82 (2000)

    MathSciNet  Article  MATH  Google Scholar 

  26. Kessler, M., Sørensen, M.: Estimating equations based on eigenfunctions for a discretely observed diffusion process. Bernoulli 5(2), 299–314 (1999)

    MathSciNet  Article  MATH  Google Scholar 

  27. Kloeden, P.E., Platen, E.: Numerical Solution of Stochastic Differential Equations, Applications of Mathematics (New York), vol. 23. Springer, Berlin (1992)

    Google Scholar 

  28. Kolmogoroff, A.: Zur Umkehrbarkeit der statistischen Naturgesetze. Math. Ann. 113(1), 766–772 (1937)

    MathSciNet  Article  MATH  Google Scholar 

  29. Larsen, K.S., Sørensen, M.: Diffusion models for exchange rates in a target zone. Math. Finance 17(2), 285–306 (2007)

    MathSciNet  Article  MATH  Google Scholar 

  30. Liu, C.S.: Ornstein–Uhlenbeck process, Cauchy process, and Ornstein–Uhlenbeck–Cauchy process on a circle. Appl. Math. Lett. 26(9), 957–962 (2013)

    MathSciNet  Article  MATH  Google Scholar 

  31. Mardia, K.V., Jupp, P.E.: Directional Statistics. Wiley Series in Probability and Statistics, 2nd edn. Wiley, Chichester (2000)

  32. Mardia, K.V.: Statistics of Directional Data, Probability and Mathematical Statistics, vol. 13. Academic Press, London (1972)

  33. Mardia, K.V.: The magic of score matching estimators and approximations for distributions on manifolds and some cutting edge applications to molecular biology. In: Proceedings 61st ISI World Statistics Congress, Marrakech (2017)

  34. Mardia, K.V., Frellsen, J.: Statistics of bivariate von Mises distributions. In: Hamelryck, T., Mardia, K.V., Ferkinghoff-Borg, J. (eds.) Bayesian Methods in Structural Bioinformatics. Statistics for Biology and Health. Springer, Berlin (2012)

    Google Scholar 

  35. Mardia, K.V., Voss, J.: Some fundamental properties of a multivariate von Mises distribution. Commun. Stat. Theory Methods 43(6), 1132–1144 (2014)

    MathSciNet  Article  MATH  Google Scholar 

  36. Mardia, K.V., Hughes, G., Taylor, C.C., Singh, H.: A multivariate von Mises distribution with applications to bioinformatics. Can. J. Stat. 36(1), 99–109 (2008)

    MathSciNet  Article  MATH  Google Scholar 

  37. McKee, S., Wall, D.P., Wilson, S.K.: An alternating direction implicit scheme for parabolic equations with mixed derivative and convective terms. J. Comput. Phys. 126(1), 64–76 (1996)

    MathSciNet  Article  MATH  Google Scholar 

  38. Øksendal, B.: Stochastic Differential Equations: An Introduction with Applications, 6th edn. Universitext, Springer, Berlin (2003)

    Google Scholar 

  39. Ozaki, T.: Statistical identification of storage models with application to stochastic hydrology. J. Am. Water Resour. Assoc. 21(4), 663–675 (1985)

    Article  Google Scholar 

  40. Papaspiliopoulos, O., Roberts, G.: Importance sampling techniques for estimation of diffusion models. In: Kessler, M., Lindner, A., Sørensen, M. (eds.) Statistical Methods for Stochastic Differential Equations. Monographs on Statistics and Applied Probability, vol. 124. Chapman & Hall/CRC Press, Boca Raton (2012)

    Google Scholar 

  41. Roberts, G.O., Stramer, O.: Langevin diffusions and Metropolis–Hastings algorithms. Methodol. Comput. Appl. Probab. 4(4), 337–357 (2003), international Workshop in Applied Probability (Caracas, 2002) (2002)

  42. Rogers, L.C.G., Williams, D.: Diffusions, Markov Processes, and Martingales. Cambridge Mathematical Library, vol. 1. Cambridge University Press, Cambridge (2000)

  43. Sermaidis, G., Papaspiliopoulos, O., Roberts, G.O., Beskos, A., Fearnhead, P.: Markov chain Monte Carlo for exact inference for diffusions. Scand. J. Stat. 40(2), 294–321 (2012)

    MathSciNet  Article  MATH  Google Scholar 

  44. Shoji, I., Ozaki, T.: A statistical method of estimation and simulation for systems of stochastic differential equations. Biometrika 85(1), 240–243 (1998)

    MathSciNet  Article  MATH  Google Scholar 

  45. Soetaert, K., Cash, J., Mazzia, F.: Solving Differential Equations in R. Use R!. Springer, New York (2012)

    Google Scholar 

  46. Sørensen, M.: Efficient estimation for ergodic diffusions sampled at high frequency. Department of Mathematical Sciences, University of Copenhagen, Technical Report (2008)

  47. Sørensen, M.: Estimating functions for diffusion-type processes. In: Kessler, M., Lindner, A., Sørensen, M. (eds.) Statistical Methods for Stochastic Differential Equations. Monographs on Statistics and Applied Probability, vol. 124. Chapman & Hall/CRC Press, Boca Raton (2012)

  48. Steele, J.M.: Stochastic Calculus and Financial Applications. Applications of Mathematics (New York), vol. 45. Springer, New York (2001)

  49. Stroock, D.W.: An Introduction to the Analysis of Paths on a Riemannian Manifold. Mathematical Surveys and Monographs, vol. 74. American Mathematical Society, Providence (2000)

  50. Thomas, J.W.: Numerical Partial Differential Equations: Finite Difference Methods. Texts in Applied Mathematics, vol. 22. Springer, New York (1995)

  51. Wehrly, T.E., Johnson, R.A.: Bivariate models for dependence of angular observations and a related Markov process. Biometrika 67(1), 255–256 (1979)

    MathSciNet  Article  MATH  Google Scholar 

  52. Yeh, S.Y., Harris, K.D.M., Jupp, P.E.: A drifting Markov process on the circle, with physical applications. Proc R Soc Lond A Mat 469(2156) (2013)

Download references

Acknowledgements

We acknowledge the insightful discussions with John Kent, Jotun Hein, and Michael Golden that led to the key motivation for the manuscript. We are grateful to Sandro Bottaro for the providing the molecular dynamics data used in the illustration. We acknowledge the valuable comments and remarks provided by two anonymous referees and an Associate Editor, which significantly improved the manuscript.

Author information

Affiliations

  1. Department of Mathematical Sciences, University of Copenhagen, Copenhagen, Denmark

    Eduardo García-Portugués & Michael Sørensen

  2. Department of Statistics, Carlos III University of Madrid, Madrid, Spain

    Eduardo García-Portugués

  3. UC3M-BS Institute of Financial Big Data, Carlos III University of Madrid, Madrid, Spain

    Eduardo García-Portugués

  4. Department of Statistics, University of Leeds, Leeds, UK

    Kanti V. Mardia

  5. Department of Statistics, University of Oxford, Oxford, UK

    Kanti V. Mardia

  6. Bioinformatics Centre, Section for Computational and RNA Biology, Department of Biology, University of Copenhagen, Copenhagen, Denmark

    Thomas Hamelryck

  7. Image Section, Department of Computer Science, University of Copenhagen, Copenhagen, Denmark

    Thomas Hamelryck

Authors
  1. Eduardo García-Portugués
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Michael Sørensen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kanti V. Mardia
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Thomas Hamelryck
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Eduardo García-Portugués.

Additional information

This work is part of the Dynamical Systems Interdisciplinary Network, University of Copenhagen. It was funded by the University of Copenhagen 2016 Excellence Programme for Interdisciplinary Research (UCPH2016-DSIN) and by Project MTM2016-76969-P from the Spanish Ministry of Economy, Industry and Competitiveness, and European Regional Development Fund (ERDF).

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

García-Portugués, E., Sørensen, M., Mardia, K.V. et al. Langevin diffusions on the torus: estimation and applications. Stat Comput 29, 1–22 (2019). https://doi.org/10.1007/s11222-017-9790-2

Download citation

Keywords

  • Circular data
  • Directional statistics
  • Likelihood
  • Protein structure
  • Stochastic Differential Equation
  • Wrapped normal

Mathematics Subject Classification

  • 60J60
  • 62M05
  • 62H11