No file available
To read the file of this research,
you can request a copy directly from the author.
Random signed measures
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
Point processes and, more generally, random measures are ubiquitous in modern statistics. However, they can only take positive values, which is a severe limitation in many situations. In this work, we introduce and study random signed measures, also known as real-valued random measures, and apply them to constrcut various Bayesian non-parametric models. In particular, we provide an existence result for random signed measures, allowing us to obtain a canonical definition for them and solve a 70-year-old open problem. Further, we provide a representation of completely random signed measures (CRSMs), which extends the celebrated Kingman's representation for completely random measures (CRMs) to the real-valued case. We then introduce specific classes of random signed measures, including the Skellam point process, which plays the role of the Poisson point process in the real-valued case, and the Gaussian random measure. We use the theoretical results to develop two Bayesian nonparametric models -- one for topic modeling and the other for random graphs -- and to investigate mean function estimation in Bayesian nonparametric regression.
ResearchGate has not been able to resolve any citations for this publication.
We consider the clustering of extremes for stationary regularly varying random fields over arbitrary growing index sets. We study sufficient assumptions on the index set such that the limit of the point processes of the exceedances above a high threshold exists. Under the so-called anti-clustering condition, the extremal dependence is only local. Thus the index set can have a general form compared to previous literature (Basrak and Planinić in Bernoulli 27(2):1371–1408, 2021; Stehr and Rønn-Nielsen in Extremes 24(4):753–795, 2021). However, we cannot describe the clustering of extreme values in terms of the usual spectral tail measure (Wu and Samorodnitsky in Stochastic Process Appl 130(7):4470–4492, 2020) except for hyperrectangles or index sets in the lattice case. Using the recent extension of the spectral measure for star-shaped equipped space (Segers et al. in Extremes 20:539–566, 2017), the Υ-spectral tail measure provides a natural extension that describes the clustering effect in full generality.
This paper reviews works on Skellam distribution, its extensions and areas of applications. Available literature shows that this distribution is flexible for modelling integer data where they appear as count data or paired count data in the field of finance, medicine, sports, and science. Bivariate Skellam distribution, dynamic Skellam model and other extensions are also discussed and additional literature are provided.
Statistical network modelling has focused on representing the graph as a discrete structure, namely the adjacency matrix. When assuming exchangeability of this array—which can aid in modelling, computations and theoretical analysis—the Aldous–Hoover theorem informs us that the graph is necessarily either dense or empty. We instead consider representing the graph as an exchangeable random measure and appeal to the Kallenberg representation theorem for this object. We explore using completely random measures (CRMs) to define the exchangeable random measure, and we show how our CRM construction enables us to achieve sparse graphs while maintaining the attractive properties of exchangeability. We relate the sparsity of the graph to the Lévy measure defining the CRM. For a specific choice of CRM, our graphs can be tuned from dense to sparse on the basis of a single parameter. We present a scalable Hamiltonian Monte Carlo algorithm for posterior inference, which we use to analyse network properties in a range of real data sets, including networks with hundreds of thousands of nodes and millions of edges.
We study intraday stochastic volatility for four liquid stocks traded on the New York Stock Exchange using a new dynamic Skellam model for high-frequency tick-by-tick discrete price changes. Since the likelihood function is analytically intractable, we rely on numerical methods for its evaluation. Given the high number of observations per series per day (1,000 to 10,000), we adopt computationally efficient methods including Monte Carlo integration. The intraday dynamics of volatility and the high number of trades without price impact require non-trivial adjustments to the basic dynamic Skellam model. In-sample residual diagnostics and goodness-of-fit statistics show that the final model provides a good fit to the data. An extensive day-to-day forecasting study of intraday volatility shows that the dynamic modified Skellam model provides accurate forecasts compared to alternative modeling approaches.
A signed network is a network with each link associated with a positive or negative sign. Models for nodes interacting over such signed networks, where two types of interactions are defined along the positive and negative links, respectively, arise from various biological, social, political, and economical systems. As modifications to the conventional DeGroot dynamics for positive links, two basic types of negative interactions along negative links, namely the opposing rule and the repelling rule, have been proposed and extensively studied. This paper reviews some fundamental convergence results that appeared recently in the literature for dynamics over deterministic or random signed networks under a unified algebraic-graphical method. We show that a systematic tool of studying node state evolution over signed networks can be obtained utilizing generalized Perron-Frobenius theory, graph theory, and elementary algebraic recursions.
Completely random measures (CRMs) and their normalizations are a rich source of Bayesian nonparametric (BNP) priors. Examples include the beta, gamma, and Dirichlet processes. In this paper we detail three classes of sequential CRM representations that can be used for simulation and posterior inference. These representations subsume existing ones that have previously been developed in an ad hoc manner for specific processes. Since a complete infinite-dimensional CRM cannot be used explicitly for computation, sequential representations are often truncated for tractability. We provide truncation error analyses for each type of sequential representation, as well as their normalized versions, thereby generalizing and improving upon existing truncation error bounds in the literature. We analyze the computational complexity of the sequential representations, which in conjunction with our error bounds allows us to study which representations are the most efficient. We include numerous applications of our theoretical results to popular (normalized) CRMs, demonstrating that our results provide a suite of tools that allow for the straightforward representation and analysis of CRMs that have not previously been used in a Bayesian nonparametric context.
We demonstrate how to calculate posteriors for general CRM-based priors and
likelihoods for Bayesian nonparametric models. We further show how to represent
Bayesian nonparametric priors as a sequence of finite draws using a
size-biasing approach---and how to represent full Bayesian nonparametric models
via finite marginals. Motivated by conjugate priors based on exponential family
representations of likelihoods, we introduce a notion of exponential families
for CRMs, which we call exponential CRMs. This construction allows us to
specify automatic Bayesian nonparametric conjugate priors for exponential CRM
likelihoods. We demonstrate that our exponential CRMs allow particularly
straightforward recipes for size-biased and marginal representations of
Bayesian nonparametric models. Along the way, we prove that the gamma process
is a conjugate prior for the Poisson likelihood process and the beta prime
process is a conjugate prior for a process we call the odds Bernoulli process.
We deliver a size-biased representation of the gamma process and a marginal
representation of the gamma process coupled with a Poisson likelihood process.
This paper describes a new class of dependent random measures which we call
compound random measure and the use of normalized versions of these random
measures as priors in Bayesian nonparametric mixture models. Their tractability
allows the properties of both compound random measures and normalized compound
random measures to be derived. In particular, we show how compound random
measures can be constructed with gamma, stable and generalized gamma process
marginals. We also derive several forms of the Laplace exponent and
characterize dependence through both the Levy copula and correlation function.
A slice sampling algorithm for inference when a normalized compound random
measure is used as the mixing measure in a nonparametric mixture model is
described and the algorithm is applied to a data example.
The study of the measurable space of signed Radon measures on a metric space is carried, establishing results of characterization of distributions of signed random measures which generalize similar results about the nonnegative case. These results enable the study of a Levy{Khintchine type characterization for inflnite divisible signed random measures.
We develop computational procedures for a class of Bayesian nonparametric and semiparametric multiplicative intensity models incorporating kernel mixtures of spatial weighted gamma measures. A key feature of our approach is that explicit expressions for posterior distributions of these models share many common structural features with the posterior distributions of Bayesian hierarchical models using the Dirichlet process. Using this fact, along with an approximation for the weighted gamma process, we show that with some care, one can adapt ef cient algorithms used for the Dirichlet process to this setting. We discuss blocked Gibbs sampling procedures and Pólya urn Gibbs samplers. We illustrate our methods with applications to proportional hazard models, Poisson spatial regression models, recurrent events, and panel count data.
This article describes a new class of prior distributions for nonparametric
function estimation. The unknown function is modeled as a limit of weighted
sums of kernels or generator functions indexed by continuous parameters that
control local and global features such as their translation, dilation,
modulation and shape. Lévy random fields and their stochastic integrals are
employed to induce prior distributions for the unknown functions or,
equivalently, for the number of kernels and for the parameters governing their
features. Scaling, shape, and other features of the generating functions are
location-specific to allow quite different function properties in different
parts of the space, as with wavelet bases and other methods employing
overcomplete dictionaries. We provide conditions under which the stochastic
expansions converge in specified Besov or Sobolev norms. Under a Gaussian error
model, this may be viewed as a sparse regression problem, with regularization
induced via the Lévy random field prior distribution. Posterior inference
for the unknown functions is based on a reversible jump Markov chain Monte
Carlo algorithm. We compare the Lévy Adaptive Regression Kernel (LARK)
method to wavelet-based methods using some of the standard test functions, and
illustrate its flexibility and adaptability in nonstationary applications.
An important issue in survival analysis is the investigation and the modeling of hazard rates. Within a Bayesian nonparametric framework, a natural and popular approach is to model hazard rates as kernel mixtures with respect to a completely random measure. In this paper we provide a comprehensive analysis of the asymptotic behavior of such models. We investigate consistency of the posterior distribution and derive fixed sample size central limit theorems for both linear and quadratic functionals of the posterior hazard rate. The general results are then specialized to various specific kernels and mixing measures yielding consistency under minimal conditions and neat central limit theorems for the distribution of functionals.
Bayesian nonparametric inference is a relatively young area of research and it has recently undergone a strong development. Most of its success can be explained by the considerable degree of exibility it ensures in statistical modelling, if compared to parametric alternatives, and by the emergence of new and ecient simulation techniques that make nonparametric models amenable to concrete use in a number of applied statistical problems. Since its introduction in 1973 by T.S. Ferguson, the Dirichlet process has emerged as a cornerstone in Bayesian nonparametrics. Nonetheless, in some cases of interest for statistical applications the Dirichlet process is not an adequate prior choice and alternative nonparametric models need to be devised. In this paper we provide a review of Bayesian nonparametric models that go beyond the Dirichlet process.
Over the past few years, interest has increased in models defined on positive and negative integers. Several application areas lead to data that are differences between positive integers. Some important examples are price changes measured discretely in financial applications, pre- and posttreatment measurements of discrete outcomes in clinical trials, the difference in the number of goals in sports events, and differencing of count-valued time series. This review aims at bringing together a wide range of models that have appeared in the literature in recent decades. We provide an extensive review on discrete distributions defined for integer data and then consider univariate and multivariate time-series models, including the class of autoregressive models, stochastic processes, and ARCH-GARCH– (autoregressive conditionally heteroskedastic–generalized autoregressive conditionally heteroskedastic–) type models.
Expected final online publication date for the Annual Review of Statistics and Its Application, Volume 10 is March 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
This book aims to present a comprehensive, self-contained, and concise overview of extreme value theory for time series, incorporating the latest research trends alongside classical methodology. Appropriate for graduate coursework or professional reference, the book requires a background in extreme value theory for i.i.d. data and basics of time series. Following a brief review of foundational concepts, it progresses linearly through topics in limit theorems and time series models while including historical insights at each chapter’s conclusion. Additionally, the book incorporates complete proofs and exercises with solutions as well as substantive reference lists and appendices, featuring a novel commentary on the theory of vague convergence.
We prove a strong law of large numbers for random signed measures on Euclidean space that holds uniformly over a family of arguments (sets) scaled by diagonal matrices. Applications to random measures generated by sums of random variables, marked point processes and stochastic integrals are also presented.
Abstract: We propose a novel statistical model for sparse networks with overlapping community structure. The model is based on representing the graph as an exchangeable point process, and naturally generalizes existing probabilistic models with overlapping block-structure to the sparse regime. Our construction builds on vectors of completely random measures, and has interpretable parameters, each node being assigned a vector representing its level of affiliation to some latent communities. We develop methods for simulating this class of random graphs, as well as to perform posterior inference. We show that the proposed approach can recover interpretable structure from two real-world networks and can handle graphs with thousands of nodes and tens of thousands of edges.
Offering the first comprehensive treatment of the theory of random measures, this book has a very broad scope, ranging from basic properties of Poisson and related processes to the modern theories of convergence, stationarity, Palm measures, conditioning, and compensation. The three large final chapters focus on applications within the areas of stochastic geometry, excursion theory, and branching processes. Although this theory plays a fundamental role in most areas of modern probability, much of it, including the most basic material, has previously been available only in scores of journal articles. The book is primarily directed towards researchers and advanced graduate students in stochastic processes and related areas.
With the rapid growth of social media, sentiment analysis, also called opinion mining, has become one of the most active research areas in natural language processing. Its application is also widespread, from business services to political campaigns. This article gives an introduction to this important area and presents some recent developments.
In this article, we present some specific aspects of symmetric Gamma process mixtures for use in regression models. First we propose a new Gibbs sampler for simulating the posterior. The algorithm is tested on two examples, the mean regression problem with normal errors, and the reconstruction of two dimensional CT images. In a second time, we establish posterior rates of convergence related to the mean regression problem with normal errors. For location-scale and location-modulation mixtures the rates are adaptive over Hölder classes, and in the case of location-modulation mixtures are nearly optimal.
Mixture models for hazard rate functions are widely used tools for addressing the statistical analysis of survival data subject to a censoring mechanism. The present article introduced a new class of vectors of random hazard rate functions that are expressed as kernel mixtures of dependent completely random measures. This leads to define dependent nonparametric prior processes that are suitably tailored to draw inferences in the presence of heterogenous observations. Besides its flexibility, an important appealing feature of our proposal is analytical tractability: we are, indeed, able to determine some relevant distributional properties and a posterior characterization that is also the key for devising an efficient Markov chain Monte Carlo sampler. For illustrative purposes, we specialize our general results to a class of dependent extended gamma processes. We finally display a few numerical examples, including both simulated and real two-sample datasets: these allow us to identify the effect of a borrowing strength phenomenon and provide evidence of the effectiveness of the prior to deal with datasets for which the proportional hazards assumption does not hold true. Supplementary materials for this article are available online.
We consider the problem of estimating the graph associated with a binary Ising Markov random field. We describe a method based on ℓ1-regularized logistic regression, in which the neighborhood of any given node is estimated by performing logistic regression subject to an ℓ1-constraint. The method is analyzed under high-dimensional scaling in which both the number of nodes p and maximum neighborhood size d are allowed to grow as a function of the number of observations n. Our main results provide sufficient conditions on the triple (n, p, d) and the model parameters for the method to succeed in consistently estimating the neighborhood of every node in the graph simultaneously. With coherence conditions imposed on the population Fisher information matrix, we prove that consistent neighborhood selection can be obtained for sample sizes n=Ω(d3log p) with exponentially decaying error. When these same conditions are imposed directly on the sample matrices, we show that a reduced sample size of n=Ω(d2log p) suffices for the method to estimate neighborhoods consistently. Although this paper focuses on the binary graphical models, we indicate how a generalization of the method of the paper would apply to general discrete Markov random fields.
Motivated by features of low latency data in financial econometrics we study in detail integer-valued Lévy processes as the basis of price processes for high frequency econometrics. We propose using models built out of the difference of two subordinators. We apply these models in practice to low latency data for a variety of different types of futures contracts.
A Gaussian random measure is a mean zero Gaussian process [eta](A), indexed by sets A in a [sigma]-field, such that [eta]([Sigma]Ai)=[Sigma][eta](Ai), where [Sigma]Ai indicates disjoint union and the series on the right is required to converge everywhere, so [eta] is a random signed measure. (This is in contrast to so-called second order random measures, which only require quadratic mean convergence.) The covariance kernel of [eta] is the signed bimeasure [nu]0(A,B)=E[eta](A)[eta](B). We give a characterization of those bimeasures which are covariance kernels of Gaussian random measures, and we show that every Gaussian random measure has an exponentially integrable total variation and is a.s. absolutely continuous with respect to a fixed finite measure on the state space.
Let μ
n
be a sequence of random finite signed measures on the locally compact group G equal to either
\mathbbTd\mathbb{T}^{d}
or ℝ
d
. We give weak conditions on the sequence μ
n
and on functions K such that the convolution product μ
n
*K, and its derivatives, converge in law, in probability, or almost surely in the Banach spaces
C0(G)\mathsf{C}_{0}(G)
or L
p
(G). Examples for sequences μ
n
covered are the empirical process (possibly arising from dependent data) and also random signed measures
Ön([(\mathbbP)\tilde]n-\mathbbP)\sqrt{n}(\tilde{\mathbb{P}}_{n}-\mathbb{P})
where
[(\mathbbP)\tilde]n\tilde{\mathbb{P}}_{n}
is some (nonparametric) estimator for the measure ℙ, including the usual kernel and wavelet based density estimators with
MISE-optimal bandwidths. As a statistical application, we apply the results to study convolutions of density estimators.
Extreme values of a stationary, multivariate time series may exhibit dependence across coordinates and over time. The aim of this paper is to offer a new and potentially useful tool called tail process to describe and model such extremes. The key property is the following fact: existence of the tail process is equivalent to multivariate regular variation of finite cuts of the original process. Certain remarkable properties of the tail process are exploited to shed new light on known results on certain point processes of extremes. The theory is shown to be applicable with great ease to stationary solutions of stochastic autoregressive processes with random coefficient matrices, an interesting special case being a recently proposed factor GARCH model. In this class of models, the distribution of the tail process is calculated by a combination of analytical methods and a novel sampling algorithm.
Completely random signed measures are defined, characterized and related to Lvy random measures and Lvy bases.
Incluye bibliografía e indice
We consider the problem of determining the distribution of means of random probability measures which are obtained by normalizing increasing additive processes. A solution is found by resorting to a well-known inversion formula for characteristic functions due to Gurland. Moreover, expressions of the posterior distributions of those means, in the presence of exchangeable observations, are given. Finally, a section is devoted to the illustration of two examples of statistical relevance.
The Bayes estimation of hazard rates for a family of multiplicative point processes is considered. We study the model for which a hazard rate can be linearly parametrized by a freely varied measure. The weighted gamma process is assumed to be the prior distribution of this measure; the posterior distributions and the posterior means are given in explicit forms. Examples of the evaluation of posterior means are given.
One of the main research areas in Bayesian Nonparametrics is the proposal and study of priors which generalize the Dirichlet process. In this paper, we provide a comprehensive Bayesian non-parametric analysis of random probabilities which are obtained by normalizing random measures with independent increments (NRMI). Special cases of these priors have already shown to be useful for statistical applications such as mixture models and species sampling problems. However, in order to fully exploit these priors, the derivation of the posterior distribution of NRMIs is crucial: here we achieve this goal and, indeed, provide explicit and tractable expressions suitable for practical implementation. The posterior distribution of an NRMI turns out to be a mixture with respect to the distribution of a specific latent variable. The analysis is completed by the derivation of the corresponding predictive distributions and by a thorough investigation of the marginal structure. These results allow to derive a generalized Blackwell-MacQueen sampling scheme, which is then adapted to cover also mixture models driven by general NRMIs. Copyright (c) 2008 Board of the Foundation of the Scandinavian Journal of Statistics.
A popular Bayesian nonparametric approach to survival analysis consists in modeling hazard rates as kernel mixtures driven by a completely random measure. In this paper we derive asymptotic results for linear and quadratic functionals of such random hazard rates. In particular, we prove central limit theorems for the cumulative hazard function and for the path-second moment and path-variance of the hazard rate. Our techniques are based on recently established criteria for the weak convergence of single and double stochastic integrals with respect to Poisson random measures. We illustrate our results by considering specific models involving kernels and random measures commonly exploited in practice.
We consider the asymptotic behavior of posterior distributions and Bayes estimators based on observations which are required to be neither independent nor identically distributed. We give general results on the rate of convergence of the posterior measure relative to distances derived from a testing criterion. We then specialize our results to independent, nonidentically distributed observations, Markov processes, stationary Gaussian time series and the white noise model. We apply our general results to several examples of infinite-dimensional statistical models including nonparametric regression with normal errors, binary regression, Poisson regression, an interval censoring model, Whittle estimation of the spectral density of a time series and a nonlinear autoregressive model. Comment: Published at http://dx.doi.org/10.1214/009053606000001172 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
+ and ξ − are two independent Poisson processes. The statement for η follows
- Kallenberg
Kallenberg (2017) ξ + and ξ − are two independent Poisson processes. The statement for η follows
Random measure priors in bayesian recovery from sketches
- M Beraha
- S Favaro
- M Sesia
Beraha, M., S. Favaro, and M. Sesia (2024). Random measure priors in bayesian recovery from
sketches. ArXiv: 2303.15029.