On Particle Methods for Parameter Estimation in State-Space Models
arXiv:1412.8695 · doi:10.1214/14-STS511
Abstract
Nonlinear non-Gaussian state-space models are ubiquitous in statistics, econometrics, information engineering and signal processing. Particle methods, also known as Sequential Monte Carlo (SMC) methods, provide reliable numerical approximations to the associated state inference problems. However, in most applications, the state-space model of interest also depends on unknown static parameters that need to be estimated from the data. In this context, standard particle methods fail and it is necessary to rely on more sophisticated algorithms. The aim of this paper is to present a comprehensive review of particle methods that have been proposed to perform static parameter estimation in state-space models. We discuss the advantages and limitations of these methods and illustrate their performance on simple models.
Published at http://dx.doi.org/10.1214/14-STS511 in the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (5)
Cited by in corpus (34)
- A framework for quantifying the value of vibration-based structural health monitoring
- Supervised learning from noisy observations: Combining machine-learning techniques with data assimilation
- Particle Filters and Data Assimilation
- Data Assimilation: The Schrödinger Perspective
- Improved and Scalable Online Learning of Spatial Concepts and Language Models with Mapping
- Sequential Monte Carlo Methods for System Identification
- Probabilistic learning of nonlinear dynamical systems using sequential Monte Carlo
- Nonlinear System Identification: Learning while respecting physical models using a sequential Monte Carlo method
- On off-line and on-line Bayesian filtering for uncertainty quantification of structural deterioration
- Latent Gaussian Count Time Series
- A probabilistic scheme for joint parameter estimation and state prediction in complex dynamical systems
- Inference of Regulatory Networks Through Temporally Sparse Data
- Approximate Smoothing and Parameter Estimation in High-Dimensional State-Space Models
- Coupling stochastic EM and Approximate Bayesian Computation for parameter inference in state-space models
- Sparse Bayesian Estimation of Parameters in Linear-Gaussian State-Space Models
- Learning of state-space models with highly informative observations: a tempered Sequential Monte Carlo solution
- Joint Online Parameter Estimation and Optimal Sensor Placement for the Partially Observed Stochastic Advection-Diffusion Equation
- Bellman filtering and smoothing for state-space models
- Fast and numerically stable particle-based online additive smoothing: the AdaSmooth algorithm
- Online Particle Smoothing with Application to Map-matching
- Online Smoothing for Diffusion Processes Observed with Noise
- Online Sequential Monte Carlo smoother for partially observed stochastic differential equations
- Improving the particle filter in high dimensions using conjugate artificial process noise
- Fisher information matrix for single molecules with stochastic trajectories
- Validating Sequential Monte Carlo for Gravitational-Wave Inference
- When Artificial Parameter Evolution Gets Real: Particle Filtering for Time-Varying Parameter Estimation in Deterministic Dynamical Systems
- Sequential Monte Carlo Squared for online inference in stochastic epidemic models
- A State-Space Perspective on Modelling and Inference for Online Skill Rating
- Warped Dynamic Linear Models for Time Series of Counts
- Likelihood-free stochastic approximation EM for inference in complex models
- Sequential Empirical Bayes method for filtering dynamic spatiotemporal processes
- Parameter Estimation for the Reduced Fracture Model via a Direct Filter Method
- Fourier Series-Based Approximation of Time-Varying Parameters in Ordinary Differential Equations
- A Point Mass Proposal Method for Bayesian State-Space Model Fitting