A Spatially Correlated Model with Generalized Autoregressive Conditionally Heteroskedastic Structure for Counts of Crimes

Crime is a negative phenomenon that affects the daily life of the population and its development. When modeling crime data, assumptions on either the spatial or the temporal relationship between observations are necessary if any statistical analysis is to be performed. In this paper, we structure space–time dependency for count data by considering a stochastic difference equation for the intensity of the space–time process rather than placing structure on a latent space–time process, as Cox processes would do. We introduce a class of spatially correlated self-exciting spatio-temporal models for count data that capture both dependence due to self-excitation, as well as dependence in an underlying spatial process. We follow the principles in Clark and Dixon (2021) but considering a generalized additive structure on spatio-temporal varying covariates. A Bayesian framework is proposed for inference of model parameters. We analyze three distinct crime datasets in the city of Riobamba (Ecuador). Our model fits the data well and provides better predictions than other alternatives.     

Modeling electricity spot and futures price dependence: A multifrequency approach

Electricity prices are known to exhibit multifractal properties. We accommodate this finding by investigating multifractal models for electricity prices. In this paper we propose a flexible Copula-MSM (Markov Switching Multifractal) approach for modeling spot and weekly futures price dynamics. By using a conditional copula function, the framework allows us to separately model the dependence structure, while enabling use of multifractal stochastic volatility models to characterize fluctuations in marginal returns. An empirical experiment is carried out using data from Nord Pool. A study of volatility forecasting performance for electricity spot prices reveals that multifractal techniques are a competitive alternative to GARCH models. We also demonstrate how the Copula-MSM model can be employed for finding optimal portfolios, which minimizes the Conditional Value-at-Risk.     

Robust Estimation for Bivariate Poisson INGARCH Models

In the integer-valued generalized autoregressive conditional heteroscedastic (INGARCH) models, parameter estimation is conventionally based on the conditional maximum likelihood estimator (CMLE). However, because the CMLE is sensitive to outliers, we consider a robust estimation method for bivariate Poisson INGARCH models while using the minimum density power divergence estimator. We demonstrate the proposed estimator is consistent and asymptotically normal under certain regularity conditions. Monte Carlo simulations are conducted to evaluate the performance of the estimator in the presence of outliers. Finally, a real data analysis using monthly count series of crimes in New South Wales and an artificial data example are provided as an illustration.     

Message Passing-Based Inference for Time-Varying Autoregressive Models

Time-varying autoregressive (TVAR) models are widely used for modeling of non-stationary signals. Unfortunately, online joint adaptation of both states and parameters in these models remains a challenge. In this paper, we represent the TVAR model by a factor graph and solve the inference problem by automated message passing-based inference for states and parameters. We derive structured variational update rules for a composite “AR node” with probabilistic observations that can be used as a plug-in module in hierarchical models, for example, to model the time-varying behavior of the hyper-parameters of a time-varying AR model. Our method includes tracking of variational free energy (FE) as a Bayesian measure of TVAR model performance. The proposed methods are verified on a synthetic data set and validated on real-world data from temperature modeling and speech enhancement tasks.     

Generalized Poisson Hurdle Model for Count Data and Its Application in Ear Disease

For count data, though a zero-inflated model can work perfectly well with an excess of zeroes and the generalized Poisson model can tackle over- or under-dispersion, most models cannot simultaneously deal with both zero-inflated or zero-deflated data and over- or under-dispersion. Ear diseases are important in healthcare, and falls into this kind of count data. This paper introduces a generalized Poisson Hurdle model that work with count data of both too many/few zeroes and a sample variance not equal to the mean. To estimate parameters, we use the generalized method of moments. In addition, the asymptotic normality and efficiency of these estimators are established. Moreover, this model is applied to ear disease using data gained from the New South Wales Health Research Council in 1990. This model performs better than both the generalized Poisson model and the Hurdle model.     

Conditional velocity statistics in the double scalar mixing layer – A mapping closure approach

In this work we use 3D direct numerical simulations (DNS) to investigate the average velocity conditioned on a conserved scalar in a double scalar mixing layer (DSML). The DSML is a canonical multistream flow designed as a model problem for the extensively studied piloted diffusion flames. The conditional mean velocity appears as an unclosed term in advanced Eulerian models of turbulent non-premixed combustion, like the conditional moment closure and transported probability density function (PDF) methods. Here it accounts for inhomogeneous effects that have been found significant in flames with relatively low Damköhler numbers. Today there are only a few simple models available for the conditional mean velocity and these are discussed with reference to the DNS results. We find that both the linear model of Kutznetzov and the Li and Bilger model are unsuitable for multi stream flows, whereas the gradient diffusion model of Pope shows very close agreement with DNS over the whole range of the DSML. The gradient diffusion model relies on a model for the conserved scalar PDF and here we have used a presumed mapping function PDF, that is known to give an excellent representation of the DNS. A new model for the conditional mean velocity is suggested by arguing that the Gaussian reference field represents the velocity field, a statement that is evidenced by a near perfect agreement with DNS. The model still suffers from an inconsistency with the unconditional flux of conserved scalar variance, though, and a strategy for developing fully consistent models is suggested.     

Statistical Ensembles for Economic Networks

Economic networks share with other social networks the fundamental property of sparsity. It is well known that the maximum entropy techniques usually employed to estimate or simulate weighted networks produce unrealistic dense topologies. At the same time, strengths should not be neglected, since they are related to core economic variables like supply and demand. To overcome this limitation, the exponential Bosonic model has been previously extended in order to obtain ensembles where the average degree and strength sequences are simultaneously fixed (conditional geometric model). In this paper a new exponential model, which is the network equivalent of Boltzmann ideal systems, is introduced and then extended to the case of joint degree-strength constraints (conditional Poisson model). Finally, the fitness of these alternative models is tested against a number of networks. While the conditional geometric model generally provides a better goodness-of-fit in terms of log-likelihoods, the conditional Poisson model could nevertheless be preferred whenever it provides a higher similarity with original data. If we are interested instead only in topological properties, the simple Bernoulli model appears to be preferable to the correlated topologies of the two more complex models.     

A New Extension of Thinning-Based Integer-Valued Autoregressive Models for Count Data

The thinning operators play an important role in the analysis of integer-valued autoregressive models, and the most widely used is the binomial thinning. Inspired by the theory about extended Pascal triangles, a new thinning operator named extended binomial is introduced, which is a general case of the binomial thinning. Compared to the binomial thinning operator, the extended binomial thinning operator has two parameters and is more flexible in modeling. Based on the proposed operator, a new integer-valued autoregressive model is introduced, which can accurately and flexibly capture the dispersed features of counting time series. Two-step conditional least squares (CLS) estimation is investigated for the innovation-free case and the conditional maximum likelihood estimation is also discussed. We have also obtained the asymptotic property of the two-step CLS estimator. Finally, three overdispersed or underdispersed real data sets are considered to illustrate a superior performance of the proposed model.     

Computational study of the mechanism of Bcl-2 apoptotic switch

In spite of attention devoted to molecular mechanisms of apoptosis, the details of functioning of one crucial component–the Bcl-2 apoptotic switch–are not completely understood. There are two competing mechanisms of its internal working—the indirect activation and the direct activation. In the absence of conclusive experimental data, we have used computational modeling to assess the properties of both mechanisms and their suitability to act as a biological switch. Since the two mechanisms form opposite poles of continuum of Bcl-2 molecular interaction models, we have constructed more general models including these two models as extreme cases. By studying the relationship between model parameters and the steady-state response we have found optimal interaction patterns which reproduce the behavior of the Bcl-2 apoptotic switch. Our results show, that stimulus–response ultrasensitivity is negatively affected by spontaneous activation of Bcl-2 effectors. We found that ultrasensitivity requires effectors activation, mediated by another subgroup of Bcl-2 proteins—activators. We have shown that the auto-activation of monomeric effector forms provides an ultrasensitivity enhancing feedback loop. Thorough robustness analysis revealed that the interaction pattern postulated in the direct activation hypothesis is able to conserve stimulus–response switching characteristics for wide range changes of its internal parameters. The robustness of the switch against the variation of the reaction parameter is strongly reduced for the intermediate hybrid model and even more for the indirect part of the models. Computer simulations of the more general model presented here suggest, that stimulus–response ultrasensitivity is an emergent property of the direct activation model that is unlikely to occur in the model of indirect activation. Introduction of indirect-model-specific interactions does not provide a better explanation of the Bcl-2 switch functionality compared to the direct model.     

Data reduction by applying an image-based modeling and rendering technique to CG models

We applied an image-based modeling and rendering technique to reduce the data size of a CG model to be obtained as a visualization result. To date, this technique has been applied to the reconstruction of 3D graphical models from real objects; the size of the models can be reduced effectively because both color information and geometric information can be reduced. We applied a silhouette-and-voxel method that does not require design data for geometric simplification. Using our system, we simplified two models, one of which, involving medical data, was reduced by about 85% in file size. An accompanying subjective quality test showed that our approach maintains approximately the same visual quality as the geometric simplification method traditionally used.     

Opinion evolution analysis for short-range and long-range Deffuant–Weisbuch models

In this paper, we propose and then analyze two generalized Deffuant–Weisbuch (DW) models. The generalized models extend the conventional DW model by taking multiple choices in two different ways. First, we demonstrate the almost sure convergence of the agent opinions for the short-range multi-choice DW dynamics when only the opinions within confidence regions may be count in. Then we analyze dynamical behavior about the long-range multi-choice DW model when some opinions out of the confidence ranges are considered with a weighted combination. Moreover, both theoretical and simulation results show that the dynamical behaviors of the two models are totally different.     

On the modeling of synchronized flow in cellular automaton models

In this paper,we further analyze our cellular automaton(CA)traffic flow model.By changing some parameters,the characteristics of our model can be significantly varied,ranging from the features of phase transitions to the number of traffic phases.We also review the other CA models based on Kerner’s three-phase traffic theory.By comparisons,we find that the core concepts for modeling the synchronized flow in these models are similar.Our model can be a good candidate for modeling the synchronized flow,since there is enough flexibility in our framework.     

A Conditional Curie–Weiss Model for Stylized Multi-group Binary Choice with Social Interaction

This paper proposes a conditional Curie–Weiss model as a model for decision making in a stylized society made up of binary decision makers that face a particular dichotomous choice between two options. Following Brock and Durlauf (Discrete choice with social interaction I: theory, 1955), we set-up both socio-economic and statistical mechanical models for the choice problem. We point out when both the socio-economic and statistical mechanical models give rise to the same self-consistent equilibrium mean choice level(s). Phase diagram of the associated statistical mechanical model and its socio-economic implications are discussed.     

New approaches in agent-based modeling of complex financial systems

Agent-based modeling is a powerful simulation technique to understand the collective behavior and microscopic interaction in complex financial systems. Recently, the concept for determining the key parameters of agent-based models from empirical data instead of setting them artificially was suggested. We first review several agent-based models and the new approaches to determine the key model parameters from historical market data. Based on the agents’ behaviors with heterogeneous personal preferences and interactions, these models are successful in explaining the microscopic origination of the temporal and spatial correlations of financial markets. We then present a novel paradigm combining big-data analysis with agent-based modeling. Specifically, from internet query and stock market data, we extract the information driving forces and develop an agent-based model to simulate the dynamic behaviors of complex financial systems.     

Random graph models for dynamic networks

Recent theoretical work on the modeling of network structure has focused primarily on networks that are static and unchanging, but many real-world networks change their structure over time. There exist natural generalizations to the dynamic case of many static network models, including the classic random graph, the configuration model, and the stochastic block model, where one assumes that the appearance and disappearance of edges are governed by continuous-time Markov processes with rate parameters that can depend on properties of the nodes. Here we give an introduction to this class of models, showing for instance how one can compute their equilibrium properties. We also demonstrate their use in data analysis and statistical inference, giving efficient algorithms for fitting them to observed network data using the method of maximum likelihood. This allows us, for example, to estimate the time constants of network evolution or infer community structure from temporal network data using cues embedded both in the probabilities over time that node pairs are connected by edges and in the characteristic dynamics of edge appearance and disappearance. We illustrate these methods with a selection of applications, both to computer-generated test networks and real-world examples.     

Sequentially Estimating the Approximate Conditional Mean Using Extreme Learning Machines

This study examined the extreme learning machine (ELM) applied to the Wald test statistic for the model specification of the conditional mean, which we call the WELM testing procedure. The omnibus test statistics available in the literature weakly converge to a Gaussian stochastic process under the null that the model is correct, and this makes their application inconvenient. By contrast, the WELM testing procedure is straightforwardly applicable when detecting model misspecification. We applied the WELM testing procedure to the sequential testing procedure formed by a set of polynomial models and estimate an approximate conditional expectation. We then conducted extensive Monte Carlo experiments to evaluate the performance of the sequential WELM testing procedure and verify that it consistently estimates the most parsimonious conditional mean when the set of polynomial models contains a correctly specified model. Otherwise, it consistently rejects all the models in the set.     

A study about the existence of the leverage effect in stochastic volatility models

The empirical relationship between the return of an asset and the volatility of the asset has been well documented in the financial literature. Named the leverage effect or sometimes risk-premium effect, it is observed in real data that, when the return of the asset decreases, the volatility increases and vice versa.Consequently, it is important to demonstrate that any formulated model for the asset price is capable of generating this effect observed in practice. Furthermore, we need to understand the conditions on the parameters present in the model that guarantee the apparition of the leverage effect.In this paper we analyze two general specifications of stochastic volatility models and their capability of generating the perceived leverage effect. We derive conditions for the apparition of leverage effect in both of these stochastic volatility models. We exemplify using stochastic volatility models used in practice and we explicitly state the conditions for the existence of the leverage effect in these examples.     

More Causes Less Effect: Destructive Interference in Decision Making

We present a new experiment demonstrating destructive interference in customers’ estimates of conditional probabilities of product failure. We take the perspective of a manufacturer of consumer products and consider two situations of cause and effect. Whereas, individually, the effect of the causes is similar, it is observed that when combined, the two causes produce the opposite effect. Such negative interference of two or more product features may be exploited for better modeling of the cognitive processes taking place in customers’ minds. Doing so can enhance the likelihood that a manufacturer will be able to design a better product, or a feature within it. Quantum probability has been used to explain some commonly observed “non-classical” effects, such as the disjunction effect, question order effect, violation of the sure-thing principle, and the Machina and Ellsberg paradoxes. In this work, we present results from a survey on the impact of multiple observed symptoms on the drivability of a vehicle. The symptoms are assumed to be conditionally independent. We demonstrate that the response statistics cannot be directly explained using classical probability, but quantum formulation easily models it, as it allows for both positive and negative “interference” between events. Since quantum formalism also accounts for classical probability’s predictions, it serves as a richer paradigm for modeling decision making behavior in engineering design and behavioral economics.     

A New First-Order Integer-Valued Autoregressive Model with Bell Innovations

A Poisson distribution is commonly used as the innovation distribution for integer-valued autoregressive models, but its mean is equal to its variance, which limits flexibility, so a flexible, one-parameter, infinitely divisible Bell distribution may be a good alternative. In addition, for a parameter with a small value, the Bell distribution approaches the Poisson distribution. In this paper, we introduce a new first-order, non-negative, integer-valued autoregressive model with Bell innovations based on the binomial thinning operator. Compared with other models, the new model is not only simple but also particularly suitable for time series of counts exhibiting overdispersion. Some properties of the model are established here, such as the mean, variance, joint distribution functions, and multi-step-ahead conditional measures. Conditional least squares, Yule–Walker, and conditional maximum likelihood are used for estimating the parameters. Some simulation results are presented to access these estimates’ performances. Real data examples are provided.     

A conditional Granger causality model approach for group analysis in functional magnetic resonance imaging

Granger causality model (GCM) derived from multivariate vector autoregressive models of data has been employed to identify effective connectivity in the human brain with functional magnetic resonance imaging (fMRI) and to reveal complex temporal and spatial dynamics underlying a variety of cognitive processes. In the most recent fMRI effective connectivity measures, pair-wise GCM has commonly been applied based on single-voxel values or average values from special brain areas at the group level. Although a few novel conditional GCM methods have been proposed to quantify the connections between brain areas, our study is the first to propose a viable standardized approach for group analysis of fMRI data with GCM. To compare the effectiveness of our approach with traditional pair-wise GCM models, we applied a well-established conditional GCM to preselected time series of brain regions resulting from general linear model (GLM) and group spatial kernel independent component analysis of an fMRI data set in the temporal domain. Data sets consisting of one task-related and one resting-state fMRI were used to investigate connections among brain areas with the conditional GCM method. With the GLM-detected brain activation regions in the emotion-related cortex during the block design paradigm, the conditional GCM method was proposed to study the causality of the habituation between the left amygdala and pregenual cingulate cortex during emotion processing. For the resting-state data set, it is possible to calculate not only the effective connectivity between networks but also the heterogeneity within a single network. Our results have further shown a particular interacting pattern of default mode network that can be characterized as both afferent and efferent influences on the medial prefrontal cortex and posterior cingulate cortex. These results suggest that the conditional GCM approach based on a linear multivariate vector autoregressive model can achieve greater accuracy in detecting network connectivity than the widely used pair-wise GCM, and this group analysis methodology can be quite useful to extend the information obtainable in fMRI.