As is the case of many studies, data collected are limited and an exact value recorded only if it falls within interval range. Hence, responses can be either left, or right censored. Linear (and nonlinear) regression models routinely used to analyze these types based on normality assumptions for errors terms. However, those analyzes might not provide robust inference when questionable. In this article, we develop a Bayesian framework censored linear by replacing Gaussian random with scale...

In many studies, the data collected are subject to some upper and lower detection limits. Hence, responses either left or right censored. A complication arises when these continuous measures present heavy tails asymmetrical behavior; simultaneously. For such structures, we propose a robust-censored linear model based on scale mixtures of skew-normal (SMSN) distributions. The SMSN is an attractive class heavy-tailed densities that includes skew-normal, skew-t, skew-slash, skew-contaminated...

In acquired immunodeficiency syndrome (AIDS) studies it is quite common to observe viral load measurements collected irregularly over time. Moreover, these can be subjected some upper and/or lower detection limits depending on the quantification assays. A complication arises when continuous repeated measures have a heavy-tailed behavior. For such data structures, we propose robust structure for censored linear model based multivariate Student’s t-distribution. To compensate autocorrelation...

AbstractThe purpose of this paper is to develop diagnostics analysis for nonlinear regression models (NLMs) under scale mixtures skew-normal (SMSN) distributions introduced by Garay et al. [Nonlinear based on SMSN distributions. J. Korean Statist. Soc. 2011;40:115–124]. This novel class provides a useful generalization the symmetrical NLM [Vanegas LH, Cysneiros FJA. Assessment diagnostic procedures in models. Comput. Data Anal. 2010;54:1002–1016] since random terms cover both symmetric as...

AbstractIn recent years, there has been considerable interest in regression models based on zero-inflated distributions. These are commonly encountered many disciplines, such as medicine, public health, and environmental sciences, among others. The Poisson (ZIP) model typically considered for these types of problems. However, the ZIP can fail if non-zero counts overdispersed relation to distribution, hence negative binomial (ZINB) may be more appropriate. In this paper, we present a Bayesian...

In this work, we consider the problem of finding moments a doubly truncated member class scale mixtures skew-normal (TSMSN) distributions. We obtain general result and then use it to derive in case versions skew-normal, skew-t, skew-slash skew-contaminated normal Many properties TSMSN family are studied, inference procedures developed simulation study is performed assess procedures. Two applications also provided, one them context censored regression models another field actuarial sciences.

The heteroscedastic nonlinear regression model (HNLM) is an important tool in data modeling. In this paper we propose a HNLM considering skew scale mixtures of normal (SSMN) distributions, which allows fitting asymmetric and heavy-tailed simultaneously. Maximum likelihood (ML) estimation performed via the expectation-maximization (EM) algorithm. observed information matrix derived analytically to account for standard errors. addition, diagnostic analysis developed using case-deletion...

Environmental data are often spatially correlated and sometimes include observations below or above detection limits (i.e., censored values reported as less more than a level of detection). Existing research studies mainly concentrate on parameter estimation using Gibbs sampling, most conducted from frequentist perspective in spatial models elusive. In this paper, we propose an exact procedure to obtain the maximum‐likelihood estimates fixed effects variance components, stochastic...

In recent years, there has been a considerable interest to study count time series with dependence structure and appearance of excess zeros values. Such are commonly encountered in diverse disciplines, such as economics, financial research, environmental science, public health, among others. this paper, we propose stationary p-order integer-valued autoregressive process zero-inflated Poisson innovations, called the ZINAR(p) times model. We some its theoretical properties develop Markov chain...

Mixed-effects models, with modifications to accommodate censored observations (LMEC/NLMEC), are routinely used analyze measurements, collected irregularly over time, which often subject some upper and lower detection limits. This paper presents a likelihood-based approach for fitting LMEC/NLMEC models autoregressive of order p dependence the error term. An EM-type algorithm is developed computing maximum likelihood estimates, obtaining as byproduct standard errors fixed effects value....

The purpose of this paper is to develop a practical framework for the analysis linear mixed-effects models censored or missing data with serial correlation errors, using multivariate Student’s t-distribution, being flexible alternative use corresponding normal distribution. We propose an efficient ECM algorithm computing maximum likelihood estimates these standard errors fixed effects and function as by-product. This uses closed-form expressions at E-step, which relies on formulas mean...