Assessing influence for pharmaceutical data in zero‐inflated generalized Poisson mixed models

Xie, Fei; Wei, Bo‐Cheng; Lin, Jin-Guan

doi:10.1002/sim.3233

Cited by 19 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Famoye and Singh (2006) fit a zero-inflated Poisson model (ZIP) to investigate domestic violence. Xie, Wei, and Lin (2008) investigated pharmaceutical data using a ZIP model with a random effect. Hall (2000) used the same model to investigate repeated measures in agriculture experiments.…”

Section: Introductionmentioning

confidence: 99%

<b>Modeling citrus huanglongbing data using a zero-inflated negative binomial distribution

et al. 2016

View full text Add to dashboard Cite

ABSTRACT. Zero-inflated data from field experiments can be problematic, as these data require the use of specific statistical models during the analysis process. This study utilized the zero-inflated negative binomial (ZINB) model with the log-and logistic-link functions to describe the incidence of plants with Huanglongbing (HLB, caused by Candidatus liberibacter spp.) in commercial citrus orchards in the Northwestern Parana State, Brazil. Each orchard was evaluated at different times. The ZINB model with random effects in both link functions provided the best fit, as the inclusion of these effects accounted for variations between orchards and the numbers of diseased plants. The results of this model show that older plants exhibit a lower probability of acquiring HLB. The application of insecticides on a calendar basis or during new foliage flushes resulted in a three times larger probability of developing HLB compared with applying insecticides only when the vector was detected.Keywords: mixed model, random effect, BLUP method, EM algorithm.Modelagem de dados huanglongbing citricos usando um modelo binomial negativo inflacionado de zero RESUMO. Em diversas áreas do conhecimento, dados com excesso de zeros são encontrados com frequência. Para a análise de tais dados, é recomendado utilizar modelos que permitam uma contagem deste excesso de zero de forma adequada. Neste artigo, o modelo binomial negativo inflacionado de zero (ZINB) foi utilizado para descrever o número de plantas doentes, acometidas por Huanglongbing, em pomares comercias de laranjeiras na região noroeste do Estado do Paraná. Entretanto, deve-se levar em consideração que cada pomar foi avaliado ao longo do tempo, sendo assim, neste contexto, o modelo ZINB com efeito aleatório em ambas as funções de ligação, logarítmica e logística, apresentou melhor ajuste aos dados, pois a introdução destes efeitos consideraram as variações entre os pomares e a dependência entre número de plantas doentes. A partir deste modelo é possível perceber que as plantas mais velhas tendem a apresentar menor probabilidade de adquirir a doença. Todavia, a aplicação de inseticidas e o manejo por calendário apresentam três vezes mais chance de apresentar a doença do que o manejo somente pela presença do vetor.Palavras-chave: modelo misto, efeito aleatório, método BLUP, algoritmo EM.

show abstract

Section: Introductionmentioning

confidence: 99%

<b>Modeling citrus huanglongbing data using a zero-inflated negative binomial distribution

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Xie et al [25] discussed the example in this article based on an influence diagnostic method and detected the strong influential observation and subject. Hall and Shen [9] proposed robust estimation for zero-inflated Poisson regression.…”

Section: Discussionmentioning

confidence: 96%

Semiparametric analysis of longitudinal zero-inflated count data

Feng

Zhu

2011

Journal of Multivariate Analysis

View full text Add to dashboard Cite

a b s t r a c tIn this article, we consider a semiparametric zero-inflated Poisson mixed model that postulates a possible nonlinear relationship between the natural logarithm of the mean of the counts and a particular covariate in the longitudinal studies. A penalized loglikelihood function is proposed and Monte Carlo expectation-maximization algorithm is used to derive the estimates. Under some mild conditions, we establish the consistency and asymptotic normality of the resulting estimators. Simulation studies are carried out to investigate the finite sample performance of the proposed method. For illustration purposes, the method is applied to a data set from a pharmaceutical company where the variable of interest is the number of episodes of side effects after the patient has taken the treatments.

show abstract

“…Xie et al. () and Chen et al. (2012) applied a local influence approach for a zero‐inflated generalized Poisson‐mixed model.…”

Section: Review Of General Theory For Local Influencementioning

confidence: 99%

“…In the context of zero‐inflated count models, Xie et al. () and Chen et al. () applied local influence to zero‐inflated generalized Poisson‐mixed models.…”

Section: Introductionmentioning

confidence: 99%

Local influence diagnostics for hierarchical count data models with overdispersion and excess zeros

et al. 2016

View full text Add to dashboard Cite

We consider models for hierarchical count data, subject to overdispersion and/or excess zeros. Molenberghs et al. () and Molenberghs et al. () extend the Poisson-normal generalized linear-mixed model by including gamma random effects to accommodate overdispersion. Excess zeros are handled using either a zero-inflation or a hurdle component. These models were studied by Kassahun et al. (). While flexible, they are quite elaborate in parametric specification and therefore model assessment is imperative. We derive local influence measures to detect and examine influential subjects, that is subjects who have undue influence on either the fit of the model as a whole, or on specific important sub-vectors of the parameter vector. The latter include the fixed effects for the Poisson and for the excess-zeros components, the variance components for the normal random effects, and the parameters describing gamma random effects, included to accommodate overdispersion. Interpretable influence components are derived. The method is applied to data from a longitudinal clinical trial involving patients with epileptic seizures. Even though the data were extensively analyzed in earlier work, the insight gained from the proposed diagnostics, statistically and clinically, is considerable. Possibly, a small but important subgroup of patients has been identified.

show abstract

Assessing influence for pharmaceutical data in zero‐inflated generalized Poisson mixed models

Cited by 19 publications

References 27 publications

<b>Modeling citrus huanglongbing data using a zero-inflated negative binomial distribution

<b>Modeling citrus huanglongbing data using a zero-inflated negative binomial distribution

Semiparametric analysis of longitudinal zero-inflated count data

Local influence diagnostics for hierarchical count data models with overdispersion and excess zeros

Contact Info

Product

Resources

About