Bayesian non-parametric modeling for integro-difference equations (original) (raw)
Related papers
Bayesian nonparametric spatial modeling with Dirichlet process mixing
Journal of the American Statistical …, 2005
Customary modeling for continuous point-referenced data assumes a Gaussian process which is often taken to be stationary. When such models are fitted within a Bayesian framework, the unknown parameters of the process are assumed to be random so a random Gaussian process results. Here, we propose a novel spatial Dirichlet process mixture model to produce a random spatial process which is neither Gaussian nor stationary. We first develop a spatial Dirichlet process model for spatial data and discuss its properties. Due to familiar limitations associated with direct use of Dirichlet process models, we introduce mixing through this process against a pure error process. We then examine properties of models created through such Dirichlet process mixing. In the Bayesian framework, posterior inference is implemented using Gibbs sampling as we detail. Spatial prediction raises interesting questions but can be handled. Finally, we illustrate the approach using simulated data as well as a dataset involving precipitation measurements over the Languedoc-Roussillon region in southern France.
Polynomial nonlinear spatio-temporal integro-difference equation models
Journal of Time Series Analysis, 2011
Spatio-temporal statistical models are ubiquitous in the environmental sciences. Often the underlying spatiotemporal process can be written as a hierarchical state-space process that utilizes prior understanding of the physical process to formulate models for the state transition. In situations with complicated dynamics, such as wave propagation, or population growth and dispersal models, parameterizing the transition function associated with the high-dimensional state process may prove difficult or even impossible. One approach to overcoming this difficulty is through the use of polynomial stochastic integro-difference equation (IDE) models. Here, in the context of discrete time and continuous space, complicated dynamics can be accommodated through redistribution kernels that are allowed to vary with space. To facilitate computation and dimension reduction, we consider a basisfunction-expansion representation of this model. Even in this framework there are too many parameters to estimate efficiently from a classical perspective. Therefore, we utilize the hierarchical Bayesian framework in order to implement a stochastic-search variable-selection algorithm that mitigates these estimation issues. Finally, the method is illustrated on the problem of long-lead prediction of equatorial sea surface temperature.
2009
The modelling and analysis of complex stochastic systems with increasingly large data sets, state-spaces and parameters provides major stimulus to research in Bayesian nonparametric methods and Bayesian computation. This dissertation presents advances in both nonparametric modelling and statistical computation stimulated by challenging problems of analysis in complex spatio-temporal systems and core computational issues in model fitting and model assessment. The first part of the thesis, represented by chapters 2 to 4, concerns novel, nonparametric Bayesian mixture models for spatial point processes, with advances in modelling, compusampling based marginal likelihood computation, the proposed adaptive Monte Carlo method and sequential learning approach can facilitate improved accuracy in marginal likelihood computation. The approaches are exemplified in studies of v both synthetic data examples, and in a real data analysis arising in astro-statistics. Finally, chapter 7 summarizes the dissertation and discusses possible extensions of the specific modelling and computational innovations, as well as potential future work.
Bayesian modeling and analysis for gradients in spatiotemporal processes
Biometrics, 2015
Stochastic process models are widely employed for analyzing spatiotemporal datasets in various scientific disciplines including, but not limited to, environmental monitoring, ecological systems, forestry, hydrology, meteorology, and public health. After inferring on a spatiotemporal process for a given dataset, inferential interest may turn to estimating rates of change, or gradients, over space and time. This manuscript develops fully model-based inference on spatiotemporal gradients under continuous space, continuous time settings. Our contribution is to offer, within a flexible spatiotemporal process model setting, a framework to estimate arbitrary directional gradients over space at any given timepoint, temporal derivatives at any given spatial location and, finally, mixed spatiotemporal gradients that reflect rapid change in spatial gradients over time and vice-versa. We achieve such inference without compromising on rich and flexible spatiotemporal process models and use nonse...
Bayesian Nonparametric Dynamic State Space Modeling with Circular Latent States
Journal of Statistical Theory and Practice, 2015
State space models are well-known for their versatility in modeling dynamic systems that arise in various scientific disciplines. Although parametric state space models are wellstudied, nonparametric approaches are much less explored in comparison. In this article we propose a novel Bayesian nonparametric approach to state space modeling assuming that both the observational and evolutionary functions are unknown and are varying with time; crucially, we assume that the unknown evolutionary equation describes dynamic evolution of some latent circular random variable. Based on appropriate kernel convolution of the standard Weiner process we model the time-varying observational and evolutionary functions as suitable Gaussian processes that take both linear and circular variables as arguments. Additionally, for the time-varying evolutionary function, we wrap the Gaussian process thus constructed around the unit circle to form an appropriate circular Gaussian process. We show that our process thus created satisfies desirable properties. For the purpose of inference we develop an MCMC based methodology combining Gibbs sampling and Metropolis-Hastings algorithms. Applications to a simulated dataset, a real wind speed dataset and a real ozone dataset demonstrated quite encouraging performances of our model and methodologies.
Data-Driven Spatio-Temporal Modeling Using the Integro-Difference Equation
IEEE Transactions on Signal Processing, 2000
A continuous-in-space, discrete-in-time dynamic spatio-temporal model known as the Integro-Difference Equation (IDE) model is presented in the context of data-driven modeling. A novel decomposition of the IDE is derived, leading to state-space representation that does not couple the number of states with the number of observation locations or the number of parameters. Based on this state-space model, an expectation-maximization (EM) algorithm is developed in order to jointly estimate the IDE model's spatial field and spatial mixing kernel. The resulting modeling framework is demonstrated on a set of examples.
Combining numerical model output and particulate data using Bayesian space-time modeling
Environmetrics, 2009
Over the past few years, Bayesian models for combining output from numerical models and air monitoring data have been applied to environmental data sets to improve spatial prediction. This paper develops a new hierarchical Bayesian model (HBM) for fine particulate matter (PM 2.5 ) that combines U. S. EPA Federal Reference Method (FRM) PM 2.5 monitoring data and Community Multi-scale Air Quality (CMAQ) numerical model output. The model is specified in a Bayesian framework and fitted using Markov Chain Monte Carlo (MCMC) techniques. We find that the statistical model combining monitoring data and CMAQ output provides reliable information about the true underlying PM 2.5 process over time and space. We base these conclusions on results of a validation exercise in which independent monitoring data were compared with predicted values from the HBM and predictions from a standard kriging model based solely on the monitoring data. regional sources of particulates, but also provides insight on the role of transboundary transport on U.S. air quality.
Dynamical non-Gaussian modelling of spatial processes
2021
Spatio-temporal processes in environmental applications are often assumed to follow a Gaussian model, possibly after some transformation. However, heterogeneity in space and time might have a pattern that will not be accommodated by transforming the data. In this scenario, modelling the variance laws is an appealing alternative. This work adds flexibility to the usual Multivariate Dynamic Gaussian model by defining the process as a scale mixture between a Gaussian and log-Gaussian processes. The scale is represented by a process varying smoothly over space and time which is allowed to depend on covariates. State-space equations define the dynamics over time for both mean and variance processes resulting in feasible inference and prediction. Analysis of artificial datasets show that the parameters are identifiable and simpler models are well recovered by the general proposed model. The analyses of two important environmental processes, maximum temperature and maximum ozone, illustrat...
Nonparametric Nonstationary Modeling of Spatio-Temporal Data Through State Space Approach
2014
Discrete-time spatial time series data arise routinely in meteorological and environmental studies. Inference and prediction associated with them are mostly carried out using any of the several variants of the linear state space model that are collectively called linear dynamic spatio-temporal models (LDSTMs). However, real world environmental processes are highly complex and are seldom representable by models with such simple linear structure. Hence, nonlinear dynamic spatio-temporal models (NLDSTMs) based on the idea of nonlinear observational and evolutionary equations have been proposed as an alternative. However, in that case, the caveat lies in selecting the specific form of nonlinearity from a large class of potentially appropriate nonlinear functions. Moreover, modeling by NLDSTMs requires precise knowledge about the dynamics underlying the data. In this article, we address this problem by introducing the Gaussian random functional dynamic spatio-temporal model (GRFDSTM). Unlike the LDSTMs or NLDSTMs, in GRFDSTM both the functions governing the observational and evolutionary equations are composed of Gaussian random functions. We exhibit many interesting theoretical properties of the GRFDSTM and demonstrate how model fitting and prediction can be carried out coherently in a Bayesian framework. We also conduct an extensive simulation study and apply our model to a real, SO 2 pollution data over Europe. The results are highly encouraging.
Bayesian Inference in Nonparametric Dynamic State-Space Models
We introduce state-space models where the functionals of the observational and the evolutionary equations are unknown, and treated as random functions evolving with time. Thus, our model is nonparametric and generalizes parametric state-space models, such as the extended Kalman filter. This random function approach also frees us from the restrictive assumption that the functional forms, although time-dependent, are of fixed forms. We specify Gaussian processes as priors of the random functions and exploit the "look-up table approach" of Bhattacharya (2007) to efficiently handle the dynamic structure of the model. We consider both univariate and multivariate situations, using the Markov chain Monte Carlo (MCMC) approach for studying the posterior distributions of interest. In the case of challenging multivariate situations we demonstrate that the newly developed Transformation-based MCMC (TMCMC) provides interesting and efficient alternatives to the usual proposal distributions. We illustrate our methods with simulated data sets, obtaining very encouraging results in both univariate and multivariate situations. Moreover, using our Gaussian process approach we analysed a real data set, which has also been analysed by and using the linearity assumption. Our analyses show that towards the end of the time series, the linearity assumption of the previous authors breaks down.