This accessible treatment offers the mathematical tools for describing and solving problems related to stochastic vector fields. Advanced undergraduates and graduate students will find its use of generalized functions a relatively simple method of resolving mathematical questions. It will prove a valuable reference for applied mathematicians and professionals in the fields of aerospace, chemical, civil, and nuclear engineering.
The author, Professor Emeritus of Engineering at Cornell University, starts with a survey of probability distributions and densities and proceeds to examinations of moments, characteristic functions, and the Gaussian distribution; random functions; and random processes in more dimensions. Extensive appendixes--which include information on Fourier transforms, tensors, generalized functions, and invariant theory--contribute toward making this volume mathematically self-contained.

A self-contained treatment, this text covers both theory and applications. Topics include homogeneous finite and infinite Markov chains, including those employed in the mathematical modeling of psychology and genetics; the basics of nonhomogeneous finite Markov chain theory; and a study of Markovian dependence in continuous time. 1980 edition.

This friendly guide is the companion you need to convert pure mathematics into understanding and facility with a host of probabilistic tools. The book provides a high-level view of probability and its most powerful applications. It begins with the basic rules of probability and quickly progresses to some of the most sophisticated modern techniques in use, including Kalman filters, Monte Carlo techniques, machine learning methods, Bayesian inference and stochastic processes. It draws on thirty years of experience in applying probabilistic methods to problems in computational science and engineering, and numerous practical examples illustrate where these techniques are used in the real world. Topics of discussion range from carbon dating to Wasserstein GANs, one of the most recent developments in Deep Learning. The underlying mathematics is presented in full, but clarity takes priority over complete rigour, making this text a starting reference source for researchers and a readable overview for students.

Fluctuating parameters appear in a variety of physical systems and phenomena. They typically come either as random forces/sources, or advecting velocities, or media (material) parameters, like refraction index, conductivity, diffusivity, etc. The well known example of Brownian particle suspended in fluid and subjected to random molecular bombardment laid the foundation for modern stochastic calculus and statistical physics. Other important examples include turbulent transport and diffusion of particle-tracers (pollutants), or continuous densities (''oil slicks''), wave propagation and scattering in randomly inhomogeneous media, for instance light or sound propagating in the turbulent atmosphere.
Such models naturally render to statistical description, where the input parameters and solutions are expressed by random processes and fields.
The fundamental problem of stochastic dynamics is to identify the essential characteristics of system (its state and evolution), and relate those to the input parameters of the system and initial data.
This raises a host of challenging mathematical issues. One could rarely solve such systems exactly (or approximately) in a closed analytic form, and their solutions depend in a complicated implicit manner on the initial-boundary data, forcing and system's (media) parameters . In mathematical terms such solution becomes a complicated "nonlinear functional" of random fields and processes.
Part I gives mathematical formulation for the basic physical models of transport, diffusion, propagation and develops some analytic tools.
Part II sets up and applies the techniques of variational calculus and stochastic analysis, like Fokker-Plank equation to those models, to produce exact or approximate solutions, or in worst case numeric procedures. The exposition is motivated and demonstrated with numerous examples.
Part III takes up issues for the coherent phenomena in stochastic dynamical systems, described by ordinary and partial differential equations, like wave propagation in randomly layered media (localization), turbulent advection of passive tracers (clustering).
Each chapter is appended with problems the reader to solve by himself (herself), which will be a good training for independent investigations.
.This book is translation from Russian and is completed with new principal results of recent research.
.The book develops mathematical tools of stochastic analysis, and applies them to a wide range of physical models of particles, fluids, and waves.
.Accessible to a broad audience with general background in mathematical physics, but no special expertise in stochastic analysis, wave propagation or turbulence"

This monograph addresses the problem of "real-time" curve fitting in the presence of noise, from the computational and statistical viewpoints. It examines the problem of nonlinear regression, where observations are made on a time series whose mean-value function is known except for a vector parameter. In contrast to the traditional formulation, data are imagined to arrive in temporal succession. The estimation is carried out in real time so that, at each instant, the parameter estimate fully reflects all available data.Specifically, the monograph focuses on estimator sequences of the so-called differential correction type. The term "differential correction" refers to the fact that the difference between the components of the updated and previous estimators is proportional to the difference between the current observation and the value that would be predicted by the regression function if the previous estimate were in fact the true value of the unknown vector parameter. The vector of proportionality factors (which is generally time varying and can depend upon previous estimates) is called the "gain" or "smoothing" vector.The main purpose of this research is to relate the large-sample statistical behavior of such estimates (consistency, rate of convergence, large-sample distribution theory, asymptotic efficiency) to the properties of the regression function and the choice of smoothing vectors. Furthermore, consideration is given to the tradeoff that can be effected between computational simplicity and statistical efficiency through the choice of gains.Part I deals with the special cases of an unknown scalar parameter-discussing probability-one and mean-square convergence, rates of mean-square convergence, and asymptotic distribution theory of the estimators for various choices of the smoothing sequence. Part II examines the probability-one and mean-square convergence of the estimators in the vector case for various choices of smoothing vectors. Examples are liberally sprinkled throughout the book. Indeed, the last chapter is devoted entirely to the discussion of examples at varying levels of generality.If one views the stochastic approximation literature as a study in the asymptotic behavior of solutions to a certain class of nonlinear first-order difference equations with stochastic driving terms, then the results of this monograph also serve to extend and complement many of the results in that literature, which accounts for the authors' choice of title.The book is written at the first-year graduate level, although this level of maturity is not required uniformly. Certainly the reader should understand the concept of a limit both in the deterministic and probabilistic senses (i.e., almost sure and quadratic mean convergence). This much will assure a comfortable journey through the first fourth of the book. Chapters 4 and 5 require an acquaintance with a few selected central limit theorems. A familiarity with the standard techniques of large-sample theory will also prove useful but is not essential. Part II, Chapters 6 through 9, is couched in the language of matrix algebra, but none of the "classical" results used are deep. The reader who appreciates the elementary properties of eigenvalues, eigenvectors, and matrix norms will feel at home.MIT Press Research Monograph No. 42

The computer age has spawned a whole new discipline of Computational Physics, a branch of physics known for its dynamic nature as opposed to the more traditional branches of experimental and theoretical physics. Within the field, the topic that has "inflated" the most with the rise of the computer is that of Stochastic Simulation, more colourfully known by its distinguished proponents, Fermi, von Neumann, Metropolis, Ulam and others such as the Monte Carlo method. Kevin MacKeown's book, the ideas of which had evolved from 15 years of teaching a final year undergraduate course on Computational Physics, serves to summarise in one volume the past and latest developments of the stochastic phenomena in the context of physics. The "teaching" approach follows a less conventional one in that there is no canon to be followed in the field. Instead, the topics are chosen to give a feeling for the breadth of applications of Monte Carlo methods in physics. Substantial references to research literature are also provided. This book is an essential reference to students wishing to gain a more technical interest in the subject as a way of getting quatitative answers to a problem. The level of knowledge of physics assumed corresponds to a that of a final year undergraduate student, but postgraduate students in a number of disciplines should also find the material of value.

Probability is an area of mathematics of tremendous contemporary importance across all aspects of human endeavour. This book is a compact account of the basic features of probability and random processes at the level of first and second year mathematics undergraduates and Masters' students in cognate fields. It is suitable for a first course in probability, plus a follow-up course in random processes including Markov chains. A special feature is the authors' attention to rigorous mathematics: not everything is rigorous, but the need for rigour is explained at difficult junctures. The text is enriched by simple exercises, together with problems (with very brief hints) many of which are taken from final examinations at Cambridge and Oxford. The first eight chapters form a course in basic probability, being an account of events, random variables, and distributions - discrete and continuous random variables are treated separately - together with simple versions of the law of large numbers and the central limit theorem. There is an account of moment generating functions and their applications. The following three chapters are about branching processes, random walks, and continuous-time random processes such as the Poisson process. The final chapter is a fairly extensive account of Markov chains in discrete time. This second edition develops the success of the first edition through an updated presentation, the extensive new chapter on Markov chains, and a number of new sections to ensure comprehensive coverage of the syllabi at major universities.

A systematic, self-contained treatment of the theory of stochastic differential equations in infinite dimensional spaces. Included is a discussion of Schwartz spaces of distributions in relation to probability theory and infinite dimensional stochastic analysis, as well as the random variables and stochastic processes that take values in infinite dimensional spaces.

Dieses Buch verschafft Ihnen einen UEberblick uber einige der bekanntesten Verfahren des maschinellen Lernens aus der Perspektive der mathematischen Statistik. Nach der Lekture kennen Sie die jeweils gestellten Forderungen an die Daten sowie deren Vor- und Nachteile und sind daher in der Lage, fur ein gegebenes Problem ein geeignetes Verfahren vorzuschlagen. Beweise werden nur dort ausfuhrlich dargestellt oder skizziert, wo sie einen didaktischen Mehrwert bieten - ansonsten wird auf die entsprechenden Fachartikel verwiesen. Fur die praktische Anwendung ist ein genaueres Studium des jeweiligen Verfahrens und der entsprechenden Fachliteratur noetig, zu der Sie auf Basis dieses Buchs aber schnell Zugang finden. Das Buch richtet sich an Studierende der Mathematik hoeheren Semesters, die bereits Vorkenntnisse in Wahrscheinlichkeitstheorie besitzen. Behandelt werden sowohl Methoden des Supervised Learning und Reinforcement Learning als auch des Unsupervised Learning. Der Umfang entspricht einer einsemestrigen vierstundigen Vorlesung. Die einzelnen Kapitel sind weitestgehend unabhangig voneinander lesbar, am Ende jedes Kapitels kann das erworbene Wissen anhand von UEbungsaufgaben und durch Implementierung der Verfahren uberpruft werden. Quelltexte in der Programmiersprache R stehen auf der Springer-Produktseite zum Buch zur Verfugung.

Das Lehrbuch vermittelt solides Basiswissen zu den thematischen Schwerpunkten Produktmasse, Fourier-Transformation, Transformationsformel, Konvergenzbegriffe, absolute Stetigkeit und Masse auf topologischen Raumen. Hoehepunkte sind die Herleitung des Riesz'schen Darstellungssatzes und der Beweis der Existenz und Eindeutigkeit des Haar'schen Masses. Der Band enthalt ferner mathematikhistorische Ausfluge und Kurzportrats von Mathematikern, die zum Thema des Buchs wichtige Beitrage geliefert haben, sowie zahlreiche UEbungsaufgaben zur Vertiefung des Stoffs.

Discusses replacement, repair, and inspection Offers estimation and statistical tests Covers accelerated life testing Explores warranty analysis manufacturing Includes service reliability

Dieses Buch ist eine Einfuhrung in die Variationsrechnung, die das Ziel hat, reellwertige Funktionale zu minimieren oder zu maximieren. Die Funktionale sind Integrale uber einem Intervall, weshalb die dafur zulassigen Funktionen von nur einer unabhangigen Variablen abhangen. Motiviert werden die Fragestellungen durch viele und zum Teil auch historisch bedeutsame Beispiele.
Die Theorie fuhrt in den Euler-Lagrange-Kalkul und in die Direkten Methoden der Variationsrechnung ein. Die Ausfuhrungen werden von Abbildungen begleitet, die das Verstandnis erleichtern. Zu jedem Abschnitt werden Ubungsaufgaben gestellt, deren Losungen am Ende des Buches zu finden sind.
Das Buch ist fur Vorlesungen ab dem 3. Semester geeignet. Die Hilfsmittel, welche uber die der Grundvorlesungen hinausgehen, werden im Text oder im Anhang bereitgestellt.
"

Aufbauend auf dem ersten Band, werden in diesem Buch weiterfuhrende Konzepte der Wahrscheinlichkeitstheorie ausfuhrlich und verstandlich diskutiert. Mit vielen exemplarisch durchgerechneten Aufgaben, einer Vielzahl weiterer Problemstellungen und ausfuhrlichen Loesungen bietet es dem Leser die Moeglichkeit, die eigenen Fahigkeiten standig zu erweitern und kritisch zu uberprufen und ein tieferes Verstandnis der Materie zu erlangen. Realitatsnahe Anwendungen ermoeglichen einen Ausblick in die breite Verwendbarkeit dieser Theorie.Auch in diesem Band wird auf die Entwicklung der Begriffsbildung und der mathematischen Konzepte besonderer Wert gelegt, sodass man ihre Bedeutung bei der Erzeugung wie auch standige Verbesserung von Forschungsinstrumenten fur die Untersuchung unserer Welt erleben kann. Gerichtet ist das Buch an Gymnasiasten, Studienanfanger an Hochschulen, Lehrer und Interessierte, die sich mit diesem Gebiet vertraut machen moechten.

Thanks to the driving forces of the Ito calculus and the Malliavin calculus, stochastic analysis has expanded into numerous fields including partial differential equations, physics, and mathematical finance. This book is a compact, graduate-level text that develops the two calculi in tandem, laying out a balanced toolbox for researchers and students in mathematics and mathematical finance. The book explores foundations and applications of the two calculi, including stochastic integrals and differential equations, and the distribution theory on Wiener space developed by the Japanese school of probability. Uniquely, the book then delves into the possibilities that arise by using the two flavors of calculus together. Taking a distinctive, path-space-oriented approach, this book crystallizes modern day stochastic analysis into a single volume.

We propose results of the investigation of the problem of mean square optimal estimation of linear functionals constructed from unobserved values of stationary stochastic processes. Estimates are based on observations of the processes with additive stationary noise process. The aim of the book is to develop methods for finding the optimal estimates of the functionals in the case where some observations are missing. Formulas for computing values of the mean-square errors and the spectral characteristics of the optimal linear estimates of functionals are derived in the case of spectral certainty, where the spectral densities of the processes are exactly known. The minimax robust method of estimation is applied in the case of spectral uncertainty, where the spectral densities of the processes are not known exactly while some classes of admissible spectral densities are given. The formulas that determine the least favourable spectral densities and the minimax spectral characteristics of the optimal estimates of functionals are proposed for some special classes of admissible densities.

The goal of this textbook is to introduce students to the stochastic analysis tools that play an increasing role in the probabilistic approach to optimization problems, including stochastic control and stochastic differential games. While optimal control is taught in many graduate programs in applied mathematics and operations research, the author was intrigued by the lack of coverage of the theory of stochastic differential games. This is the first title in SIAM's Financial Mathematics book series and is based on the author's lecture notes. It will be helpful to students who are interested in stochastic differential equations (forward, backward, forward-backward); the probabilistic approach to stochastic control (dynamic programming and the stochastic maximum principle); and mean field games and control of McKean-Vlasov dynamics. The theory is illustrated by applications to models of systemic risk, macroeconomic growth, flocking/schooling, crowd behavior, and predatory trading, among others.

In many real-world applications, the problems with the data used for scheduling such as processing times, set-up times, release dates or due dates is not exactly known before applying a specific solution algorithm which restricts practical aspects of scheduling theory. During the last decades, several approaches have been developed for sequencing and scheduling with inaccurate data, depending on whether the data is given as random numbers, fuzzy numbers or whether it is uncertain (ie: it can take values from a given interval). This book considers the four major approaches for dealing with such problems: a stochastic approach, a fuzzy approach, a robust approach and a stability approach. Each of the four parts is devoted to one of these approaches. First, it contains a survey chapter on this subject, as well as between further chapters, presenting some recent research results in the particular area. The book provides the reader with a comprehensive and up-to-date introduction into scheduling with inaccurate data. The four survey chapters deal with scheduling with stochastic approaches, fuzzy job-shop scheduling, min-max regret scheduling problems and a stability approach to sequencing and scheduling under uncertainty. This book will be useful for applied mathematicians, students and PhD students dealing with scheduling theory, optimisation and calendar planning.

This monograph presents important research results in the areas of queuing theory, risk theory, graph theory and reliability theory. The analysed stochastic network models are aggregated systems of elements in random environments. To construct and to analyse a large number of different stochastic network models it is possible by a proof of new analytical results and a construction of calculation algorithms besides of the application of cumbersome traditional techniques Such a constructive approach is in a prior detailed investigation of an algebraic model component and leads to an appearance of new original stochastic network models, algorithms and application to computer science and information technologies. Accuracy and asymptotic formulas, additional calculation algorithms have been constructed due to an introduction of control parameters into analysed models, a reduction of multi-dimensional problems to one dimensional problems, a comparative analysis, a graphic interpretation of network models, an investigation of new models characteristics, a choice of special distributions classes or principles of subsystems aggregation, proves of new statements.

This book presents both the fundamentals and the major research topics in statistical physics of systems out of equilibrium. It summarizes different approaches to describe such systems on the thermodynamic and stochastic levels, and discusses a variety of areas including reactions, anomalous kinetics, and the behavior of self-propelling particles.

The monograph is devoted to an investigation of co-operative effects in stochastic models. It includes original results of the authors in the last decade. The main object of the monograph is an analysis of an influence of a stochastic model structure on its characteristics. Problems of a co-operation and a decomposition are actual in a solution of a lot of concrete problems. These problems are: a parallelisation of algorithms and programs, a modelling of supercomputers, computer networks, systems of mobile telephones catastrophes in complex systems, a design and an improvement of technological and economical processes etc. The co-operative effects create a source of significant dependencies between complex system characteristics under large random disturbances. To analyse these effects is necessary to create special methods based on structural analysis of multi-element stochastic models together with majoral asymptotic bounds of these models characteristics. At the same time it demands to develop new approaches to a processing of statistical data and a skill in an usage of the probability theory limit theorems and related asymptotic series and bounds. A choice of the monograph material is defined as by initial applied problems so by probability methods of their solution. Conditionally the monograph may be divided into two parts. First of them contains four sections devoted to a finding of the co-operative effects and to a development of new related analytical and numerical methods. This part has presumably methodological character and creates a theoretical base of an investigation of applied stochastic systems. Second part contains three sections devoted to a solution of different applied problems. It has some interesting substantial results.

Focusing on one of the major branches of probability theory, this book treats the large class of processes with continuous sample paths that possess the Markov property. The exposition is based on the theory of stochastic analysis, which uses such notions as stochastic differentials and stochastic integrals.

This book is the second volume of a two-volume monograph devoted to the study of limit and ergodic theorems for regularly and singularly perturbed Markov chains, semi-Markov processes, and multi-alternating regenerative processes with semi-Markov modulation. The second volume presents a complete classification of ergodic theorems for alternating regenerative processes, including more than twenty-five such theorems. The text addresses new asymptotic recurrent algorithms of phase space reduction for multi-alternating regenerative processes modulating by regularly and singularly perturbed finite semi-Markov processes. It also features a new study of super-long, long, and short time ergodic theorems for these processes. The book also contains a comprehensive bibliography of major works in the field. It provides an effective reference for both graduate students as well as theoretical and applied researchers studying stochastic processes and their applications.

Stochastic Modeling for Medical Image Analysis provides a brief introduction to medical imaging, stochastic modeling, and model-guided image analysis. Today, image-guided computer-assisted diagnostics (CAD) faces two basic challenging problems. The first is the computationally feasible and accurate modeling of images from different modalities to obtain clinically useful information. The second is the accurate and fast inferring of meaningful and clinically valid CAD decisions and/or predictions on the basis of model-guided image analysis. To help address this, this book details original stochastic appearance and shape models with computationally feasible and efficient learning techniques for improving the performance of object detection, segmentation, alignment, and analysis in a number of important CAD applications. The book demonstrates accurate descriptions of visual appearances and shapes of the goal objects and their background to help solve a number of important and challenging CAD problems. The models focus on the first-order marginals of pixel/voxel-wise signals and second- or higher-order Markov-Gibbs random fields of these signals and/or labels of regions supporting the goal objects in the lattice. This valuable resource presents the latest state of the art in stochastic modeling for medical image analysis while incorporating fully tested experimental results throughout.

An introduction to general theories of stochastic processes and modern martingale theory. The volume focuses on consistency, stability and contractivity under geometric invariance in numerical analysis, and discusses problems related to implementation, simulation, variable step size algorithms, and random number generation.