Statistical Inference and Applications of a Spatial-temporal Markov Random Field

Statistical Inference and Applications of a Spatial-temporal Markov Random Field

Author: Zack Nadrich

Publisher:

Published: 2020

Total Pages: 51

ISBN-13:

DOWNLOAD EBOOK

Markov random fields (MRF) form a broad class of stochastic models frequently applied to spatial data. A generalization of Markov chains, the MRF models capture spatial correlation by introducing dependence among data on the surface through a chosen neighborhood structure. We introduce a three-dimensional MRF with such neighborhoods that incorporate spatial patterns as well as the time dimension to create a spatio-temporal model. The proposed clique configuration forces spatial dependencies to evolve through time, reflecting dynamics of an observed process.Our statistical inference approach to the Markov Random field is likelihood based. The complex form of the joint distribution of observed spatially and longitudinally dependent data does not allow a closed form of the likelihood function. We show that the pseudolikelihood of this MRF model, as applied to Bernoulli data, can be conveniently expressed as logistic regression. The theory of maximum pseudolikelihood estimation shows that our resulting parameter estimates are consistent and asymptotically normal. As a case study, we use our Markov random field specification to model the dynamics and spread of wildfires. We show that the model can be used to detect wildfire spread and explain the direction and speed at which a wildfire is moving, as well as changes in their behavior in time and in space. We also apply the Markov random field as a generative model in simulations to develop accurate, timely, and probabilistic wildfire spread forecasts, to complement state-of-the-art physical models.


Gaussian Markov Random Fields

Gaussian Markov Random Fields

Author: Havard Rue

Publisher: CRC Press

Published: 2005-02-18

Total Pages: 280

ISBN-13: 0203492021

DOWNLOAD EBOOK

Gaussian Markov Random Field (GMRF) models are most widely used in spatial statistics - a very active area of research in which few up-to-date reference works are available. This is the first book on the subject that provides a unified framework of GMRFs with particular emphasis on the computational aspects. This book includes extensive case-studie


Random Fields for Spatial Data Modeling

Random Fields for Spatial Data Modeling

Author: Dionissios T. Hristopulos

Publisher: Springer Nature

Published: 2020-02-17

Total Pages: 884

ISBN-13: 9402419187

DOWNLOAD EBOOK

This book provides an inter-disciplinary introduction to the theory of random fields and its applications. Spatial models and spatial data analysis are integral parts of many scientific and engineering disciplines. Random fields provide a general theoretical framework for the development of spatial models and their applications in data analysis. The contents of the book include topics from classical statistics and random field theory (regression models, Gaussian random fields, stationarity, correlation functions) spatial statistics (variogram estimation, model inference, kriging-based prediction) and statistical physics (fractals, Ising model, simulated annealing, maximum entropy, functional integral representations, perturbation and variational methods). The book also explores links between random fields, Gaussian processes and neural networks used in machine learning. Connections with applied mathematics are highlighted by means of models based on stochastic partial differential equations. An interlude on autoregressive time series provides useful lower-dimensional analogies and a connection with the classical linear harmonic oscillator. Other chapters focus on non-Gaussian random fields and stochastic simulation methods. The book also presents results based on the author’s research on Spartan random fields that were inspired by statistical field theories originating in physics. The equivalence of the one-dimensional Spartan random field model with the classical, linear, damped harmonic oscillator driven by white noise is highlighted. Ideas with potentially significant computational gains for the processing of big spatial data are presented and discussed. The final chapter concludes with a description of the Karhunen-Loève expansion of the Spartan model. The book will appeal to engineers, physicists, and geoscientists whose research involves spatial models or spatial data analysis. Anyone with background in probability and statistics can read at least parts of the book. Some chapters will be easier to understand by readers familiar with differential equations and Fourier transforms.


Spatial Applications of Markov Random Fields and Neural Networks for Spatio-temporal Denoising, Causal Inference and Reinforcement Learning

Spatial Applications of Markov Random Fields and Neural Networks for Spatio-temporal Denoising, Causal Inference and Reinforcement Learning

Author: Mauricio Benjamín García Tec

Publisher:

Published: 2022

Total Pages: 0

ISBN-13:

DOWNLOAD EBOOK

Discrete spatial structures are ubiquitous in statistical analysis. They can take the form of images, grids, and more generally, graphs. This work develops novel methodology leading to broadly applicable algorithms of graph smoothing and neural newtorks to improve statistical learning in a variety of tasks and spatially-structured domains, including temporal and sequential decision-making processes. Thus, each chapter corresponds to a case study with applications in spatio-temporal denoising, causal inference, and reinforcement learning. Graph smoothing methods are used in all of them and their effectiveness is evaluated. In addition, some chapters develop more specialized methods that further exploit the spatial and statistical structure of the data. One of the objectives sustained throughout the work will be developing scalable algorithms to handle high-resolution spatial data or other computationally demanding scenarios


Statistical Inference in Stochastic Processes

Statistical Inference in Stochastic Processes

Author: N.U. Prabhu

Publisher: CRC Press

Published: 1990-12-18

Total Pages: 294

ISBN-13: 9780824784171

DOWNLOAD EBOOK

Covering both theory and applications, this collection of eleven contributed papers surveys the role of probabilistic models and statistical techniques in image analysis and processing, develops likelihood methods for inference about parameters that determine the drift and the jump mechanism of a di


Statistical Methods for Spatio-Temporal Systems

Statistical Methods for Spatio-Temporal Systems

Author: Barbel Finkenstadt

Publisher: Chapman and Hall/CRC

Published: 2006-10-20

Total Pages: 286

ISBN-13: 9781584885931

DOWNLOAD EBOOK

Statistical Methods for Spatio-Temporal Systems presents current statistical research issues on spatio-temporal data modeling and will promote advances in research and a greater understanding between the mechanistic and the statistical modeling communities. Contributed by leading researchers in the field, each self-contained chapter starts with an introduction of the topic and progresses to recent research results. Presenting specific examples of epidemic data of bovine tuberculosis, gastroenteric disease, and the U.K. foot-and-mouth outbreak, the first chapter uses stochastic models, such as point process models, to provide the probabilistic backbone that facilitates statistical inference from data. The next chapter discusses the critical issue of modeling random growth objects in diverse biological systems, such as bacteria colonies, tumors, and plant populations. The subsequent chapter examines data transformation tools using examples from ecology and air quality data, followed by a chapter on space-time covariance functions. The contributors then describe stochastic and statistical models that are used to generate simulated rainfall sequences for hydrological use, such as flood risk assessment. The final chapter explores Gaussian Markov random field specifications and Bayesian computational inference via Gibbs sampling and Markov chain Monte Carlo, illustrating the methods with a variety of data examples, such as temperature surfaces, dioxin concentrations, ozone concentrations, and a well-established deterministic dynamical weather model.


Statistical Methods for Spatio-Temporal Systems

Statistical Methods for Spatio-Temporal Systems

Author: Barbel Finkenstadt

Publisher: CRC Press

Published: 2006-10-20

Total Pages: 314

ISBN-13: 1420011057

DOWNLOAD EBOOK

Statistical Methods for Spatio-Temporal Systems presents current statistical research issues on spatio-temporal data modeling and will promote advances in research and a greater understanding between the mechanistic and the statistical modeling communities. Contributed by leading researchers in the field, each self-contained chapter starts w


Scalable Bayesian spatial analysis with Gaussian Markov random fields

Scalable Bayesian spatial analysis with Gaussian Markov random fields

Author: Per Sidén

Publisher: Linköping University Electronic Press

Published: 2020-08-17

Total Pages: 53

ISBN-13: 9179298184

DOWNLOAD EBOOK

Accurate statistical analysis of spatial data is important in many applications. Failing to properly account for spatial autocorrelation may often lead to false conclusions. At the same time, the ever-increasing sizes of spatial datasets pose a great computational challenge, as many standard methods for spatial analysis are limited to a few thousand data points. In this thesis, we explore how Gaussian Markov random fields (GMRFs) can be used for scalable analysis of spatial data. GMRFs are closely connected to the commonly used Gaussian processes, but have sparsity properties that make them computationally cheap both in time and memory. The Bayesian framework enables a GMRF to be used as a spatial prior, comprising the assumption of smooth variation over space, and gives a principled way to estimate the parameters and propagate uncertainty. We develop new algorithms that enable applying GMRF priors in 3D to the brain activity inherent in functional magnetic resonance imaging (fMRI) data, with millions of observations. We show that our methods are both faster and more accurate than previous work. A method for approximating selected elements of the inverse precision matrix (i.e. the covariance matrix) is also proposed, which is important for evaluating the posterior uncertainty. In addition, we establish a link between GMRFs and deep convolutional neural networks, which have been successfully used in countless machine learning tasks for images, resulting in a deep GMRF model. Finally, we show how GMRFs can be used in real-time robotic search and rescue operations, for modeling the spatial distribution of injured persons. Tillförlitlig statistisk analys av spatiala data är viktigt inom många tillämpningar. Om inte korrekt hänsyn tas till spatial autokorrelation kan det ofta leda till felaktiga slutsatser. Samtidigt ökar ständigt storleken på de spatiala datamaterialen vilket utgör en stor beräkningsmässig utmaning, eftersom många standardmetoder för spatial analys är begränsade till några tusental datapunkter. I denna avhandling utforskar vi hur Gaussiska Markov-fält (eng: Gaussian Markov random fields, GMRF) kan användas för mer skalbara analyser av spatiala data. GMRF-modeller är nära besläktade med de ofta använda Gaussiska processerna, men har gleshetsegenskaper som gör dem beräkningsmässigt effektiva både vad gäller tids- och minnesåtgång. Det Bayesianska synsättet gör det möjligt att använda GMRF som en spatial prior som innefattar antagandet om långsam spatial variation och ger ett principiellt tillvägagångssätt för att skatta parametrar och propagera osäkerhet. Vi utvecklar nya algoritmer som gör det möjligt att använda GMRF-priors i 3D för den hjärnaktivitet som indirekt kan observeras i hjärnbilder framtagna med tekniken fMRI, som innehåller milliontals datapunkter. Vi visar att våra metoder är både snabbare och mer korrekta än tidigare forskning. En metod för att approximera utvalda element i den inversa precisionsmatrisen (dvs. kovariansmatrisen) framförs också, vilket är viktigt för att kunna evaluera osäkerheten i posteriorn. Vidare gör vi en koppling mellan GMRF och djupa neurala faltningsnätverk, som har använts framgångsrikt för mängder av bildrelaterade problem inom maskininlärning, vilket mynnar ut i en djup GMRF-modell. Slutligen visar vi hur GMRF kan användas i realtid av autonoma drönare för räddningsinsatser i katastrofområden för att modellera den spatiala fördelningen av skadade personer.


Spatial Statistics and Modeling

Spatial Statistics and Modeling

Author: Carlo Gaetan

Publisher: Springer Science & Business Media

Published: 2009-11-10

Total Pages: 308

ISBN-13: 0387922571

DOWNLOAD EBOOK

Spatial statistics are useful in subjects as diverse as climatology, ecology, economics, environmental and earth sciences, epidemiology, image analysis and more. This book covers the best-known spatial models for three types of spatial data: geostatistical data (stationarity, intrinsic models, variograms, spatial regression and space-time models), areal data (Gibbs-Markov fields and spatial auto-regression) and point pattern data (Poisson, Cox, Gibbs and Markov point processes). The level is relatively advanced, and the presentation concise but complete. The most important statistical methods and their asymptotic properties are described, including estimation in geostatistics, autocorrelation and second-order statistics, maximum likelihood methods, approximate inference using the pseudo-likelihood or Monte-Carlo simulations, statistics for point processes and Bayesian hierarchical models. A chapter is devoted to Markov Chain Monte Carlo simulation (Gibbs sampler, Metropolis-Hastings algorithms and exact simulation). A large number of real examples are studied with R, and each chapter ends with a set of theoretical and applied exercises. While a foundation in probability and mathematical statistics is assumed, three appendices introduce some necessary background. The book is accessible to senior undergraduate students with a solid math background and Ph.D. students in statistics. Furthermore, experienced statisticians and researchers in the above-mentioned fields will find the book valuable as a mathematically sound reference. This book is the English translation of Modélisation et Statistique Spatiales published by Springer in the series Mathématiques & Applications, a series established by Société de Mathématiques Appliquées et Industrielles (SMAI).


Statistical Inference for Piecewise-deterministic Markov Processes

Statistical Inference for Piecewise-deterministic Markov Processes

Author: Romain Azais

Publisher: John Wiley & Sons

Published: 2018-07-31

Total Pages: 279

ISBN-13: 1119544033

DOWNLOAD EBOOK

Piecewise-deterministic Markov processes form a class of stochastic models with a sizeable scope of applications: biology, insurance, neuroscience, networks, finance... Such processes are defined by a deterministic motion punctuated by random jumps at random times, and offer simple yet challenging models to study. Nevertheless, the issue of statistical estimation of the parameters ruling the jump mechanism is far from trivial. Responding to new developments in the field as well as to current research interests and needs, Statistical inference for piecewise-deterministic Markov processes offers a detailed and comprehensive survey of state-of-the-art results. It covers a wide range of general processes as well as applied models. The present book also dwells on statistics in the context of Markov chains, since piecewise-deterministic Markov processes are characterized by an embedded Markov chain corresponding to the position of the process right after the jumps.