Search

Direct marketing identifies customers who buy, more probable, a specific product to reduce the cost and increase the response rate of a marketing campaign. The advancement of technology in the current era makes the data collection process easy. Hence, a large number of customer data can be stored in companies where they can be employed to solve the direct marketing problem. In this paper, a novel Bayesian method titled correlation-augment naïve Bayes (CAN) is proposed to improve the conventional naïve Bayes (NB) classifier. The performance of the proposed method in terms of the response rate is evaluated and compared to several well-known Bayesian networks and other well-known classifiers based on seven real-world datasets from different areas with different characteristics. The experimental results show that the proposed CAN method has a much better performance compared to the other investigated methods for direct marketing in almost all cases. [ABSTRACT FROM AUTHOR]

In recent years, there has been a considerable interest to study count time series with a dependence structure and appearance of excess of zeros values. Such series are commonly encountered in diverse disciplines, such as economics, financial research, environmental science, public health, among others. In this paper, we propose a stationary p-order integer-valued autoregressive process with zero-inflated Poisson innovations, called the ZINAR(p) times series model. We study some of its theoretical properties and develop a Markov chain Monte Carlo (MCMC) algorithm for inferring parameters from Bayesian perspectives. Finally, we demonstrate the utility of proposed ZINAR(p) model through simulated and real data examples. [ABSTRACT FROM AUTHOR]

This paper considers posterior consistency in the context of high-dimensional variable selection using the Bayesian lasso algorithm. In a frequentist setting, consistency is perhaps the most basic property that we expect any reasonable estimator to achieve. However, in a Bayesian setting, consistency is often ignored or taken for granted, especially in more complex hierarchical Bayesian models. In this paper, we have derived sufficient conditions for posterior consistency in the Bayesian lasso model with the orthogonal design, where the number of parameters grows with the sample size. [ABSTRACT FROM AUTHOR]

Feature selection arises in many areas of modern science. For example, in genomic research, we want to find the genes that can be used to separate tissues of different classes (e.g. cancer and normal). One approach is to fit regression/classification models with certain penalization. In the past decade, hyper-LASSO penalization (priors) have received increasing attention in the literature. However, fully Bayesian methods that use Markov chain Monte Carlo (MCMC) for regression/classification with hyper-LASSO priors are still in lack of development. In this paper, we introduce an MCMC method for learning multinomial logistic regression with hyper-LASSO priors. Our MCMC algorithm uses Hamiltonian Monte Carlo in a restricted Gibbs sampling framework. We have used simulation studies and real data to demonstrate the superior performance of hyper-LASSO priors compared to LASSO, and to investigate the issues of choosing heaviness and scale of hyper-LASSO priors. [ABSTRACT FROM AUTHOR]

A good method of combining Bayesian networks (BNs) should be a generic one that ensures a combined BN meets three important criteria of avoiding cycles, preserving conditional independencies, and preserving the characteristics of individual BN parameters. All combination methods assumed that there is an ancestral ordering shared by individual BNs. If this assumption is violated, then avoiding cycles may be inefficient. In this paper, without considering an ancestral ordering, we introduce a novel method for aggregation of BNs. For this purpose, we first combine the BNs using the modification of the method introduced by Feng et al. We then use the simulated annealing algorithm for getting an acyclic graph in which the minimum arcs have been removed. Using this method, most of the conditional independencies are preserved. We compare the results of the proposed method with the two classical BNs combination methods; union and intersection, and hence to demonstrate the distinctive advantages of the proposed BNs combination method. [ABSTRACT FROM AUTHOR]

Bayesian techniques have been widely used in finite element model (FEM) updating. The attraction of these techniques is their ability to quantify and characterize the uncertainties associated with dynamic systems. In order to update an FEM, the Bayesian formulation requires the evaluation of the posterior distribution function. For large systems, this function is difficult to solve analytically. In such cases, the use of sampling techniques often provides a good approximation of this posterior distribution function. The hybrid Monte Carlo (HMC) method is a classic sampling method used to approximate high-dimensional complex problems. However, the acceptance rate of HMC is sensitive to the system size, as well as to the time step used to evaluate the molecular dynamics trajectory. The shadow HMC technique (SHMC), which is a modified version of the HMC method, was developed to improve sampling for large system sizes by drawing from a modified shadow Hamiltonian function. However, the SHMC algorithm performance is limited by the use of a non-separable modified Hamiltonian function. Moreover, two additional parameters are required for the sampling procedure, which could be computationally expensive. To overcome these weaknesses, the separable shadow HMC (S2HMC) method has been introduced. This method uses a transformation to a different parameter space to generate samples. In this paper, we analyse the application and performance of these algorithms, including the parameters used in each algorithm, their limitations and the effects on model updating. The accuracy and the efficiency of the algorithms are demonstrated by updating the finite element models of two real mechanical structures. It is observed that the S2HMC algorithm has a number of advantages over the other algorithms; for example, the S2HMC algorithm is able to efficiently sample at larger time steps while using fewer parameters than the other algorithms. [ABSTRACT FROM PUBLISHER]

This paper introduces a evolutionary computation method that applies Bayesian classifiers to optimization problems. This approach is based on Estimation of Distribution Algorithms (EDAs) in which Bayesian or Gaussian networks are applied to the evolution of a population of individuals (i.e. potential solutions to the optimization problem) in order to improve the quality of the individuals of the next generation. Our new approach, called Evolutionary Bayesian Classifier-based Optimization Algorithm (EBCOA), employs Bayesian classifiers instead of Bayesian or Gaussian networks in order to evolve individuals to a fitter population. In brief, EBCOAs are characterized by applying Bayesian classification techniques - usually applied to supervised classification problems - to optimization in continuous domains. We propose and review in this paper different Bayesian classifiers for implementing our EBCOA method, focusing particularly on EBCOAs applying naïve Bayes, semi-na¨ive Bayes, and tree augmented na¨ive Bayes classifiers. This work presents a deep study on the behavior of these algorithms with classical optimiztion problems in continuous domains. The different parameters used for tuning the performance of the algorithms are discussed, and a comprehensive overview of their influence is provided. We also present experimental results to compare this new method with other state of the art approaches of the evolutionary computation field for continuous domains such as Evolutionary Strategies (ES) and Estimation of Distribution Algorithms (EDAs). [ABSTRACT FROM AUTHOR]

This paper presents a comparative study of several well-known and thoroughly tested techniques for the segmentation of textured images, including two algorithms belonging to the adaptive Bayesian family of restoration and segmentation methods, and a novel approach based on the recently introduced concept of the frequency histogram of connected elements (FHCE). The paper first introduces the parameters that define a connected element and then details the sensitivity analysis of these parameters, showing that the grayscale intensity histogram of a digital image is a particular case of the FHCE. The application domain chosen for comparison purposes is the problem of medical images segmentation and, more specifically, as a particularly illustrative case the segmentation of digital angiograms is analyzed in detail. To get a comparative evaluation of FHCE performance, two well-established adaptive or contextual Bayesian segmentation algorithms have been applied to the segmentation of digital angiograms as well. The paper ends with a brief discussion of the comparative performances. [ABSTRACT FROM AUTHOR]

This paper presents a Bayesian analysis of partially linear additive models for quantile regression. We develop a semiparametric Bayesian approach to quantile regression models using a spectral representation of the nonparametric regression functions and the Dirichlet process (DP) mixture for error distribution. We also consider Bayesian variable selection procedures for both parametric and nonparametric components in a partially linear additive model structure based on the Bayesian shrinkage priors via a stochastic search algorithm. Based on the proposed Bayesian semiparametric additive quantile regression model referred to as BSAQ, the Bayesian inference is considered for estimation and model selection. For the posterior computation, we design a simple and efficient Gibbs sampler based on a location-scale mixture of exponential and normal distributions for an asymmetric Laplace distribution, which facilitates the commonly used collapsed Gibbs sampling algorithms for the DP mixture models. Additionally, we discuss the asymptotic property of the sempiparametric quantile regression model in terms of consistency of posterior distribution. Simulation studies and real data application examples illustrate the proposed method and compare it with Bayesian quantile regression methods in the literature. [ABSTRACT FROM AUTHOR]

The full Bayesian significance test (FBST) was introduced by Pereira and Stern for measuring the evidence of a precise null hypothesis. The FBST requires both numerical optimization and multidimensional integration, whose computational cost may be heavy when testing a precise null hypothesis on a scalar parameter of interest in the presence of a large number of nuisance parameters. In this paper we propose a higher order approximation of the measure of evidence for the FBST, based on tail area expansions of the marginal posterior of the parameter of interest. When in particular focus is on matching priors, further results are highlighted. Numerical illustrations are discussed. [ABSTRACT FROM PUBLISHER]

In this paper, a Bayesian network is developed to investigate three intertwining parking decisions, namely parking period, parking location, and parking duration, and the impacts of a number of parking-related factors on these decisions. With parking information from Beijing, China in 2005, the structure and parameter of a Bayesian network were learnt by employing the K2 algorithm and Bayesian parameter estimation method respectively. The results show that the decision on how long to park follows that on where to park, and both of them are affected by the decision of when to park. This suggests that parking policies aimed at intervening in one specific parking decision may have an indirect influence on other parking decisions, which embraces an integrated view in the development of parking policies. The findings facilitate the development of measures for regulating parking behavior by identifying important contributing factors. [ABSTRACT FROM PUBLISHER]

This paper proposes a procedure of synthetic detection for the location of a change point and outliers in bilinear time series models with a change after an unknown time point. Based on Bayesian framework, we first derive the conditional posterior distribution of the change point and from that distribution estimate the position of the change point. Then we use these results to detect the outliers in the time series before and after that change point via Gibbs sampler algorithm. Our simulation studies show that the proposed procedure is effective. [ABSTRACT FROM PUBLISHER]

Quasi-optimal procedures of testing many hypotheses are described in this paper. They significantly simplify the Bayesian algorithms of hypothesis testing and computation of the risk function. The relations allowing for obtaining the estimations for the values of average risks in optimum tasks are given. The obtained general solutions are reduced to concrete formulae for a multivariate normal distribution of probabilities. The methods of approximate computation of the risk functions in Bayesian tasks of testing many hypotheses are offered. The properties and interrelations of the developed methods and algorithms are investigated. On the basis of a simulation, the validity of the obtained results and conclusions drawn is presented. [ABSTRACT FROM AUTHOR]

Item response theory (IRT) comprises a set of statistical models which are useful in many fields, especially when there is an interest in studying latent variables (or latent traits). Usually such latent traits are assumed to be random variables and a convenient distribution is assigned to them. A very common choice for such a distribution has been the standard normal. Recently, Azevedo et al. [Bayesian inference for a skew-normal IRT model under the centred parameterization, Comput. Stat. Data Anal. 55 (2011), pp. 353–365] proposed a skew-normal distribution under the centred parameterization (SNCP) as had been studied in [R.B. Arellano-Valle and A. Azzalini, The centred parametrization for the multivariate skew-normal distribution, J. Multivariate Anal. 99(7) (2008), pp. 1362–1382], to model the latent trait distribution. This approach allows one to represent any asymmetric behaviour concerning the latent trait distribution. Also, they developed a Metropolis–Hastings within the Gibbs sampling (MHWGS) algorithm based on the density of the SNCP. They showed that the algorithm recovers all parameters properly. Their results indicated that, in the presence of asymmetry, the proposed model and the estimation algorithm perform better than the usual model and estimation methods. Our main goal in this paper is to propose another type of MHWGS algorithm based on a stochastic representation (hierarchical structure) of the SNCP studied in [N. Henze, A probabilistic representation of the skew-normal distribution, Scand. J. Statist. 13 (1986), pp. 271–275]. Our algorithm has only one Metropolis–Hastings step, in opposition to the algorithm developed by Azevedo et al., which has two such steps. This not only makes the implementation easier but also reduces the number of proposal densities to be used, which can be a problem in the implementation of MHWGS algorithms, as can be seen in [R.J. Patz and B.W. Junker, A straightforward approach to Markov Chain Monte Carlo methods for item response models, J. Educ. Behav. Stat. 24(2) (1999), pp. 146–178; R.J. Patz and B.W. Junker, The applications and extensions of MCMC in IRT: Multiple item types, missing data, and rated responses, J. Educ. Behav. Stat. 24(4) (1999), pp. 342–366; A. Gelman, G.O. Roberts, and W.R. Gilks, Efficient Metropolis jumping rules, Bayesian Stat. 5 (1996), pp. 599–607]. Moreover, we consider a modified beta prior (which generalizes the one considered in [3]) and a Jeffreys prior for the asymmetry parameter. Furthermore, we study the sensitivity of such priors as well as the use of different kernel densities for this parameter. Finally, we assess the impact of the number of examinees, number of items and the asymmetry level on the parameter recovery. Results of the simulation study indicated that our approach performed equally as well as that in [3], in terms of parameter recovery, mainly using the Jeffreys prior. Also, they indicated that the asymmetry level has the highest impact on parameter recovery, even though it is relatively small. A real data analysis is considered jointly with the development of model fitting assessment tools. The results are compared with the ones obtained by Azevedo et al. The results indicate that using the hierarchical approach allows us to implement MCMC algorithms more easily, it facilitates diagnosis of the convergence and also it can be very useful to fit more complex skew IRT models. [ABSTRACT FROM PUBLISHER]