BAYESIAN HIDDEN MARKOV MODELS FOR LONGITUDINAL COUNTS

This paper describes a Bayesian approach to modelling carcinogenity in animal studies where the data consist of counts of the number of tumours present over time. It compares two autoregressive hidden Markov models. One of them models the transitions between three latent states: an inactive transient state, a multiplying state for increasing counts and a reducing state for decreasing counts. The second model introduces a fourth tied state to describe non‐zero observations that are neither increasing nor decreasing. Both these models can model the length of stay upon entry of a state. A discrete constant hazards waiting time distribution is used to model the time to onset of tumour growth. Our models describe between‐animal‐variability by a single hierarchy of random effects and the within‐animal variation by first‐order serial dependence. They can be extended to higher‐order serial dependence and multi‐level hierarchies. Analysis of data from animal experiments comparing the influence of two genes leads to conclusions that differ from those of Dunson (2000). The observed data likelihood defines an information criterion to assess the predictive properties of the three‐ and four‐state models. The deviance information criterion is appropriately defined for discrete parameters.     

Survival functions for the frailty models based on the discrete compound Poisson process

Frailty models are often used to model heterogeneity in survival analysis. The distribution of the frailty is generally assumed to be continuous. In some circumstances, it is appropriate to consider discrete frailty distributions. Having zero frailty can be interpreted as being immune, and population heterogeneity may be analysed using discrete frailty models. In this paper, survival functions are derived for the frailty models based on the discrete compound Poisson process. Maximum likelihood estimation procedures for the parameters are studied. We examine the fit of the models to earthquake and the traffic accidents’ data sets from Turkey.     

Bayesian diffusion process models with time-varying parameters

The diffusion process is a widely used statistical model for many natural dynamic phenomena but its inference is very complicated because complete data describing the diffusion sample path is not necessarily available. In addition, data is often collected with substantial uncertainty and it is not uncommon to have missing observations. Thus, the observed process will be discrete over a finite time period and the marginal likelihood given by this discrete data is not always available. In this paper, we consider a class of nonstationary diffusion process models with not only the measurement error but also discretely time-varying parameters which are modeled via a state space model. Hierarchical Bayesian inference for such a diffusion process model with time-varying parameters is applied to financial data.     

The Identifiability of Dependent Competing Risks Models Induced by Bivariate Frailty Models

In this paper, we propose to use a special class of bivariate frailty models to study dependent censored data. The proposed models are closely linked to Archimedean copula models. We give sufficient conditions for the identifiability of this type of competing risks models. The proposed conditions are derived based on a property shared by Archimedean copula models and satisfied by several well‐known bivariate frailty models. Compared with the models studied by Heckman and Honoré and Abbring and van den Berg, our models are more restrictive but can be identified with a discrete (even finite) covariate. Under our identifiability conditions, expectation–maximization (EM) algorithm provides us with consistent estimates of the unknown parameters. Simulation studies have shown that our estimation procedure works quite well. We fit a dependent censored leukaemia data set using the Clayton copula model and end our paper with some discussions. © 2014 Board of the Foundation of the Scandinavian Journal of Statistics     

The potential of integrated modelling in conservation biology: A case study of the black‐footed albatross (Phoebastria nigripes)

Conservation biology aims at assessing the status of a population, based on information which is often incomplete. Integrated population modelling based on state‐space models appears to be a powerful and relevant way of combining into a single likelihood several types of information such as capture‐recapture data and population surveys. In this paper, the authors describe the principles of integrated population modelling and they evaluate its performance for conservation biology based on a case study, that of the black‐footed albatross, a northern Pacific albatross species suspected to be impacted by longline fishing     

Dependence Estimation for High‐frequency Sampled Multivariate CARMA Models

The paper considers high‐frequency sampled multivariate continuous‐time autoregressive moving average (MCARMA) models and derives the asymptotic behaviour of the sample autocovariance function to a normal random matrix. Moreover, we obtain the asymptotic behaviour of the cross‐covariances between different components of the model. We will see that the limit distribution of the sample autocovariance function has a similar structure in the continuous‐time and in the discrete‐time model. As a special case, we consider a CARMA (one‐dimensional MCARMA) process. For a CARMA process, we prove Bartlett's formula for the sample autocorrelation function. Bartlett's formula has the same form in both models; only the sums in the discrete‐time model are exchanged by integrals in the continuous‐time model. Finally, we present limit results for multivariate MA processes as well, which are not known in this generality in the multivariate setting yet.     

Multistate recapture models: Modelling incomplete individual histories

Multistate capture-recapture models are a natural generalization of the usual one-site recapture models. Similarly, individuals are sampled on discrete occasions, at which they may be captured or not. However, contrary to the one-site case, the individuals can move within a finite set of states between occasions. The growing interest in spatial aspects of population dynamics presently contributes to making multistate models a very promising tool for population biology. We review first the interest and the potential of multistate models, in particular when they are used with individual states as well as geographical sites. Multistate models indeed constitute canonical capture-recapture models for individual categorical covariates changing over time, and can be linked to longitudinal studies with missing data and models such as hidden Markov chains. Multistate models also provide a promising tool for handling heterogeneity of capture, provided states related to capturability can be defined and used. Such an approach could be relevant for population size estimation in closed populations. Multistate models also constitute a natural framework for mixtures of information in individual history data. Presently, most models can be fit using program MARK. As an example, we present a canonical model for multisite accession to reproduction, which fully generalizes a classical one-site model. In the generalization proposed, one can estimate simultaneously age-dependent rates of accession to reproduction, natal and breeding dispersal. Finally, we discuss further generalizations - such as a multistate generalization of growth rate models and models for data where the state in which an individual is detected is known with uncertainty - and prospects for software development.     

Multistate recapture models: modelling incomplete individual histories

Multistate capture-recapture models are a natural generalization of the usual one-site recapture models. Similarly, individuals are sampled on discrete occasions, at which they may be captured or not. However, contrary to the one-site case, the individuals can move within a finite set of states between occasions. The growing interest in spatial aspects of population dynamics presently contributes to making multistate models a very promising tool for population biology. We review first the interest and the potential of multistate models, in particular when they are used with individual states as well as geographical sites. Multistate models indeed constitute canonical capture-recapture models for individual categorical covariates changing over time, and can be linked to longitudinal studies with missing data and models such as hidden Markov chains. Multistate models also provide a promising tool for handling heterogeneity of capture, provided states related to capturability can be defined and used. Such an approach could be relevant for population size estimation in closed populations. Multistate models also constitute a natural framework for mixtures of information in individual history data. Presently, most models can be fit using program MARK. As an example, we present a canonical model for multisite accession to reproduction, which fully generalizes a classical one-site model. In the generalization proposed, one can estimate simultaneously age-dependent rates of accession to reproduction, natal and breeding dispersal. Finally, we discuss further generalizations - such as a multistate generalization of growth rate models and models for data where the state in which an individual is detected is known with uncertainty - and prospects for software development.     

Flexible uncertainty in mixture models for ordinal responses

In classical mixture models for ordinal data with an uncertainty component, the Uniform distribution is used to model indecision. In the approach proposed here, the discrete Uniform distribution is replaced by a more flexible distribution, which is centered in the middle of the response categories. The resulting model allows to distinguish between a tendency to middle categories and a tendency to extreme categories. By linking these preferences to explanatory variables, one can investigate which persons show a tendency to these response styles. It is demonstrated that severe bias might occur if inadvertently the Uniform distribution is used to model uncertainty. An application to attitudes on the performance of health services illustrates the advantages of the more flexible model.     

Multilevel models for longitudinal data

Summary.  Repeated measures and repeated events data have a hierarchical structure which can be analysed by using multilevel models. A growth curve model is an example of a multilevel random-coefficients model, whereas a discrete time event history model for recurrent events can be fitted as a multilevel logistic regression model. The paper describes extensions to the basic growth curve model to handle auto-correlated residuals, multiple-indicator latent variables and correlated growth processes, and event history models for correlated event processes. The multilevel approach to the analysis of repeated measures data is contrasted with structural equation modelling. The methods are illustrated in analyses of children's growth, changes in social and political attitudes, and the interrelationship between partnership transitions and childbearing.     

SPACE–TIME MODELLING OF SYDNEY HARBOUR WINDS

This paper develops a space‐time statistical model for local forecasting of surface‐level wind fields in a coastal region with complex topography. The statistical model makes use of output from deterministic numerical weather prediction models which are able to produce forecasts of surface wind fields on a spatial grid. When predicting surface winds at observing stations , errors can arise due to sub‐grid scale processes not adequately captured by the numerical weather prediction model , and the statistical model attempts to correct for these influences. In particular , it uses information from observing stations within the study region as well as topographic information to account for local bias. Bayesian methods for inference are used in the model , with computations carried out using Markov chain Monte Carlo algorithms. Empirical performance of the model is described , illustrating that a structured Bayesian approach to complicated space‐time models of the type considered in this paper can be readily implemented and can lead to improvements in forecasting over traditional methods.     

Continuous-discrete state-space modeling of panel data with nonlinear filter algorithms

Continuous time models with sampled data possess several advantages over conventional discrete time series and panel models (cf., e.g. special issue Stat. Neerl. 62(1), 2008). For example, data with unequal time intervals between the waves can be treated efficiently, since the model parameters of the dynamical system model are not affected by the measurement process. The continuous-discrete state space model is a combination of continuous time dynamics (stochastic differential equations, SDE) and discrete time noisy measurements.     

SIMULATED MAXIMUM LIKELIHOOD APPLIED TO NON-GAUSSIAN AND NONLINEAR MIXED EFFECTS AND STATE–SPACE MODELS

The paper presents an overview of maximum likelihood estimation using simulated likelihood, including the use of antithetic variables and evaluation of the simulation error of the resulting estimates. It gives a general purpose implementation of simulated maximum likelihood and uses it to re‐visit four models that have previously appeared in the published literature: a state–space model for count data; a nested random effects model for binomial data; a nonlinear growth model with crossed random effects; and a crossed random effects model for binary salamander‐mating data. In the case of the last three examples, this appears to be the first time that maximum likelihood fits of these models have been presented.     

Uncertainty Quantification in Case of Imperfect Models: A Non‐Bayesian Approach

The starting point in uncertainty quantification is a stochastic model, which is fitted to a technical system in a suitable way, and prediction of uncertainty is carried out within this stochastic model. In any application, such a model will not be perfect, so any uncertainty quantification from such a model has to take into account the inadequacy of the model. In this paper, we rigorously show how the observed data of the technical system can be used to build a conservative non‐asymptotic confidence interval on quantiles related to experiments with the technical system. The construction of this confidence interval is based on concentration inequalities and order statistics. An asymptotic bound on the length of this confidence interval is presented. Here we assume that engineers use more and more of their knowledge to build models with order of errors bounded by . The results are illustrated by applying the newly proposed approach to real and simulated data.     

Multivariate Poisson hidden Markov models with a case study of modelling seismicity

While the literature on multivariate models for continuous data flourishes, there is a lack of models for multivariate counts. We aim to contribute to this framework by extending the well known class of univariate hidden Markov models to the multidimensional case, by introducing multivariate Poisson hidden Markov models. Each state of the extended model is associated with a different multivariate discrete distribution. We consider different distributions with Poisson marginals, starting from the multivariate Poisson distribution and then extending to copula based distributions to allow flexible dependence structures. An EM type algorithm is developed for maximum likelihood estimation. A real data application is presented to illustrate the usefulness of the proposed models. In particular, we apply the models to the occurrence of strong earthquakes (surface wave magnitude ≥5), in three seismogenic subregions in the broad region of the North Aegean Sea for the time period from 1 January 1981 to 31 December 2008. Earthquakes occurring in one subregion may trigger events in adjacent ones and hence the observed time series of events are cross‐correlated. It is evident from the results that the three subregions interact with each other at times differing by up to a few months. This migration of seismic activity is captured by the model as a transition to a state of higher seismicity.     

MULTILEVEL DISCRETE‐TIME EVENT HISTORY MODELS WITH APPLICATIONS TO THE ANALYSIS OF RECURRENT EMPLOYMENT TRANSITIONS

Data on the timing of events such as births, residential moves and changes in employment status are collected in many longitudinal surveys. These data often have a highly complex structure, with events of several types occurring repeatedly over time to an individual and interdependences between different event processes (e.g. births and employment transitions). The aim of this paper is to review a general class of multilevel discrete‐time event history models for handling recurrent events and transitions between multiple states. It is also shown how standard methods can be extended to allow for time‐varying covariates that are outcomes of an event process that is jointly determined with the process of interest. The considerable potential of these methods for studying transitions through the life course is illustrated in analyses of the effect of the presence and age of children on women's employment transitions, using data from the British Household Panel Survey.     

Higher Moments and Prediction‐Based Estimation for the COGARCH(1,1) Model

COGARCH models are continuous time versions of the well‐known GARCH models of financial returns. The first aim of this paper is to show how the method of prediction‐based estimating functions can be applied to draw statistical inference from observations of a COGARCH(1,1) model if the higher‐order structure of the process is clarified. A second aim of the paper is to provide recursive expressions for the joint moments of any fixed order of the process. Asymptotic results are given, and a simulation study shows that the method of prediction‐based estimating function outperforms the other available estimation methods.     

Modelling of unexpected shift in SPC

Optimal statistical process control (SPC) requires models of both in-control and out-of-control process states. Whereas a normal distribution is the generally accepted model for the in-control state, there is a doubt as to the existence of reliable models for out-of-control cases. Various process models, available in the literature, for discrete manufacturing systems (parts industry) can be treated as bounded discrete-space Markov chains, completely characterized by the original in-control state and a transition matrix for shifts to an out-of-control state. The present work extends these models by using a continuous-state Markov chain, incorporating non-random corrective actions. These actions are to be realized according to the SPC technique and should substantially affect the model. The developed stochastic model yields a Laplace distribution of a process mean. An alternative approach, based on the Information theory, also results in a Laplace distribution. Real-data tests confirm the applicability of a Laplace distribution for the parts industry and show that the distribution parameter is mainly controlled by the SPC sample size.     

EXPONENTIAL SMOOTHING AND NON‐NEGATIVE DATA

The most common forecasting methods in business are based on exponential smoothing, and the most common time series in business are inherently non‐negative. Therefore it is of interest to consider the properties of the potential stochastic models underlying exponential smoothing when applied to non‐negative data. We explore exponential smoothing state space models for non‐negative data under various assumptions about the innovations, or error, process. We first demonstrate that prediction distributions from some commonly used state space models may have an infinite variance beyond a certain forecasting horizon. For multiplicative error models that do not have this flaw, we show that sample paths will converge almost surely to zero even when the error distribution is non‐Gaussian. We propose a new model with similar properties to exponential smoothing, but which does not have these problems, and we develop some distributional properties for our new model. We then explore the implications of our results for inference, and compare the short‐term forecasting performance of the various models using data on the weekly sales of over 300 items of costume jewelry. The main findings of the research are that the Gaussian approximation is adequate for estimation and one‐step‐ahead forecasting. However, as the forecasting horizon increases, the approximate prediction intervals become increasingly problematic. When the model is to be used for simulation purposes, a suitably specified scheme must be employed.     

Graphical Network Models for International Financial Flows

The late-2000s financial crisis stressed the need to understand the world financial system as a network of countries, where cross-border financial linkages play a fundamental role in the spread of systemic risks. Financial network models, which take into account the complex interrelationships between countries, seem to be an appropriate tool in this context. To improve the statistical performance of financial network models, we propose to generate them by means of multivariate graphical models. We then introduce Bayesian graphical models, which can take model uncertainty into account, and dynamic Bayesian graphical models, which provide a convenient framework to model temporal cross-border data, decomposing the model into autoregressive and contemporaneous networks. The article shows how the application of the proposed models to the Bank of International Settlements locational banking statistics allows the identification of four distinct groups of countries, that can be considered central in systemic risk contagion.