Parameter estimation for partially observed hypoelliptic diffusions

Summary.  Hypoelliptic diffusion processes can be used to model a variety of phenomena in applications ranging from molecular dynamics to audio signal analysis. We study parameter estimation for such processes in situations where we observe some components of the solution at discrete times. Since exact likelihoods for the transition densities are typically not known, approximations are used that are expected to work well in the limit of small intersample times Δt and large total observation times N Δt. Hypoellipticity together with partial observation leads to ill conditioning requiring a judicious combination of approximate likelihoods for the various parameters to be estimated. We combine these in a deterministic scan Gibbs sampler alternating between missing data in the unobserved solution components, and parameters. Numerical experiments illustrate asymptotic consistency of the method when applied to simulated data. The paper concludes with an application of the Gibbs sampler to molecular dynamics data.

[1]  Simon J. Godsill,et al.  Bayesian Inference for Continuous-Time Arma Models Driven by Non-Gaussian LÉVY Processes , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[2]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[3]  G. Roberts,et al.  On inference for partially observed nonlinear diffusion models using the Metropolis–Hastings algorithm , 2001 .

[4]  Simon J. Godsill,et al.  Estimation of CAR processes observed in noise using Bayesian inference , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[5]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[6]  Jessica G. Gaines,et al.  Variable Step Size Control in the Numerical Solution of Stochastic Differential Equations , 1997, SIAM J. Appl. Math..

[7]  Tohru Ozaki,et al.  Comparative study of estimation methods for continuous time stochastic processes , 1997 .

[8]  Peter E. Kloeden,et al.  On effects of discretization on estimators of drift parameters for diffusion processes , 1996, Journal of Applied Probability.

[9]  R. Tweedie,et al.  Exponential convergence of Langevin distributions and their discrete approximations , 1996 .

[10]  R. Durrett.  Stochastic Calculus: A Practical Introduction , 1996 .

[11]  D. Nualart.  The Malliavin Calculus and Related Topics , 1995 .

[12]  P. Tavan,et al.  Molecular dynamics of conformational substates for a simplified protein model , 1994 .

[13]  P. Kloeden,et al.  Numerical Solution of Stochastic Differential Equations , 1992 .

[14]  D. Catlin.  Estimation, Control, and the Discrete Kalman Filter , 1988 .

[15]  B. Øksendal.  Stochastic differential equations : an introduction with applications , 1987 .

[16]  D. Florens-zmirou,et al.  Estimation of the coefficients of a diffusion from discrete observations , 1986 .

[17]  M. Karplus,et al.  CHARMM: A program for macromolecular energy, minimization, and dynamics calculations , 1983 .

[18]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[19]  R. Kálmán.  A new approach to linear filtering and prediction problems" transaction of the asme~journal of basic , 1960 .

[20]  H. Kramers.  Brownian motion in a field of force and the diffusion model of chemical reactions , 1940 .

[21]  Y Pokern,et al.  Fitting stochastic differential equations to molecular dynamics data. , 2007 .

[22]  Physikalische Gesellschaft.  Position-dependent diffusion coefficients and free energies from Bayesian analysis of equilibrium and replica molecular dynamics simulations , 2005 .

[23]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[24]  Jun S. Liu,et al.  Monte Carlo strategies in scientific computing , 2001 .

[25]  R. Khasminskii,et al.  On the estimation of parameters for linear stochastic differential equations , 1999 .

[26]  INVERSE PROBLEMS NEWSLETTER , 1997 .

[27]  R. Tweedie,et al.  Exponential Convergence of Langevin Diiusions and Their Discrete Approximations , 1997 .

[28]  L. Trefethen,et al.  Numerical linear algebra , 1997 .

[29]  Xiao-Li Meng,et al.  The EM Algorithm—an Old Folk‐song Sung to a Fast New Tune , 1997 .

[30]  Bo Martin Bibby,et al.  On Estimation for Discretely Observed Diffusions: A Review , 1996 .

[31]  P. Kloeden,et al.  Numerical Solutions of Stochastic Differential Equations , 1995 .

[32]  A. Pedersen.  A new approach to maximum likelihood estimation for stochastic differential equations based on discrete observations , 1995 .

[33]  Michael C. Mackey,et al.  Chaos, Fractals, and Noise , 1994 .

[34]  Jean Jacod,et al.  On the estimation of the diffusion coefficient for multi-dimensional diffusion processes , 1993 .

[35]  D. Florens-zmirou.  Approximate discrete-time schemes for statistics of diffusion processes , 1989 .

[36]  M. Musiela,et al.  Some parameter estimation problems for hypoelliptic homogeneous Gaussian diffusions , 1985 .

[37]  C. Gardiner.  Handbook of Stochastic Methods , 1983 .