Paulo Justiniano Ribeiro Junior

Paulo Justiniano Ribeiro Junior

Auto-generated profile, based on publicly available information

Paulo Justiniano Ribeiro Junior

Universidade Federal do Paraná, Brazil

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 June 2014 in Journal of Applied Statistics
Verified
Journal Article
Open access
WZWalmes Marques Zeviani
Walmes Marques Zeviani
PRPaulo Justiniano Ribeiro
Paulo Justiniano Ribeiro
WBWagner Hugo Bonat
Wagner Hugo Bonat
SSSilvia Emiko Shimakura
Silvia Emiko Shimakura
JMJoel Augusto Muniz
Joel Augusto Muniz
Event counts are response variables with non-negative integer values representing the number of times that an event occurs within a fixed domain such as a time interval, a geographical area or a cell of a contingency table. Analysis of counts by Gaussian regression models ignores the discreteness, asymmetry and heteroscedasticity and is inefficient, providing unrealistic standard errors or possibly negative predictions of the expected number of events. The Poisson regression is the standard model for count data with underlying assumptions on the generating process which may be implausible in many applications. Statisticians have long recognized the limitation of imposing equidispersion under the Poisson regression model. A typical situation is when the conditional variance exceeds the conditional mean, in which case models allowing for overdispersion are routinely used. Less reported is the case of underdispersion with fewer modeling alternatives and assessments available in the literature. One of such alternatives, the Gamma-count model, is adopted here in the analysis of an agronomic experiment designed to investigate the effect of levels of defoliation on different phenological states upon the number of cotton bolls. Data set and code for analysis are available as online supplements. Results show improvements over the Poisson model and the semi-parametric quasi-Poisson model in capturing the observed variability in the data. Estimating rather than assuming the underlying variance process leads to important insights into the process.

21 December 2018 in Communications in Statistics - Theory and Methods
Verified
Journal Article
Open access
ARAna Julia Righetto
Ana Julia Righetto
CFChristel Faes
Christel Faes
YVYannick Vandendijck
Yannick Vandendijck
PRPaulo Justiniano Ribeiro
Paulo Justiniano Ribeiro
Many methods used in spatial statistics are computationally demanding, and so, the development of more computationally efficient methods has received attention. A important development is the integrated nested Laplace approximation method which is carry out Bayesian analysis more efficiently This method, for geostatistical data, is done considering the SPDE approach that requires the creation of a mesh overlying the study area and all the obtained results depend on it. The impact of the mesh on inference and prediction is investigated through simulations. As there is no formal procedure to specify it, we investigate a guideline to create an optimal mesh.

1 April 2013 in Revista Brasileira de Ciência do Solo
Verified
Journal Article
Open access
PDF file
CNCristiano Nunes Nesi
Cristiano Nunes Nesi
Andressa RibeiroAndressa Ribeiro
Andressa Ribeiro
WBWagner Hugo Bonat
Wagner Hugo Bonat
PJPaulo Justiniano Ribeiro Junior
Paulo Justiniano Ribeiro Junior
Na modelagem estatística da variabilidade espacial, estimam-se os parâmetros da dependência espacial, que são utilizados na interpolação de valores em locais não amostrados. Para tal, o processo de modelagem deve ser realizado com critérios estatísticos que garantam predições confiáveis e representem a real variabilidade local. Este trabalho avaliou diferentes formulações do modelo geoestatístico gaussiano para reconstituir a superfície que representa o fósforo (P) na área, a partir de medições dos teores de P em 48 parcelas experimentais localizadas em Xanxerê, SC, destacando o método utilizado nas análises. A combinação da presença de covariáveis no modelo e a necessidade de transformação para normalidade dos dados definiram quatro alternativas para modelagem. Utilizou-se a função de correlação de Matèrn, avaliada nos valores 0,5; 1,5; e 2,5 para parâmetro de suavidade. Os modelos foram comparados pelo valor maximizado do logaritmo da função de verossimilhança e também por validação cruzada. O modelo selecionado foi o que incorporou a variável resposta transformada, as coordenadas da área como covariáveis e o valor 0,5 para o parâmetro de suavidade. As medidas de validação cruzada pouco acrescentaram aos resultados de comparação por verossimilhança, que evidenciaram que na modelagem geoestatística, o cuidado com observações globais ou locais atípicas, além da seleção com base em diferentes modelos, deve ser o foco para obter resultados compatíveis com a realidade.

1 June 2013 in Idesia (Arica)
Verified
Journal Article
Open access
PDF file
CNCristiano Nunes Nesi
Cristiano Nunes Nesi
SSSilvia Emiko Shimakura
Silvia Emiko Shimakura
PJPaulo Justiniano Ribeiro Junior
Paulo Justiniano Ribeiro Junior
LMLouise Larissa May De Mio
Louise Larissa May De Mio
Survival analysis is applied when the time until the occurrence of an event is the object of interest. Such data are routinely collected in plant pathology whereas application of methods of survival analysis is uncommon. Basic concepts in survival analysis for use in plant pathology such as Kaplan-Meier methods, log-rank test and Cox proportional hazards models are introduced and applied estimating the effects of cultivars on the survival times of brown rot symptoms and on the instantaneous risk of expressing symptoms in a hypothetical study.

2 March 2023 in Journal of Statistical Computation and Simulation
Verified
Journal Article
Closed access
GSGuilherme Parreira da Silva
Guilherme Parreira da Silva
HLHenrique Aparecido Laureano
Henrique Aparecido Laureano
RPRicardo Rasmussen Petterle
Ricardo Rasmussen Petterle
PJPaulo Justiniano Ribeiro Jr
Paulo Justiniano Ribeiro Jr
WBWagner Hugo Bonat
Wagner Hugo Bonat
Researchers are often interested in understanding the relationship between a set of covariates and a set of response variables. To achieve this goal, the use of regression analysis, either linear or generalized linear models, is largely applied. However, such models only allow users to model one response variable at a time. Moreover, it is not possible to directly calculate from the regression model a correlation measure between the response variables. In this article, we employed the Multivariate Generalized Linear Mixed Models framework, which allows the specification of a set of response variables and calculates the correlation between them through a random effect structure that follows a multivariate normal distribution. We used the maximum likelihood estimation framework to estimate all model parameters using Laplace approximation to integrate out the random effects. The derivatives are provided by automatic differentiation. The outer maximization was made using a general-purpose algorithm such as PORT and Broyden–Fletcher–Goldfarb–Shanno algorithm (BFGS). We delimited this problem by studying count response variables with the following distributions: Poisson, negative binomial, Conway-Maxwell-Poisson (COM-Poisson), and double Poisson. While the first distribution can model only equidispersed data, the second models equi and overdispersed, and the third and fourth models all types of dispersion (i.e. including underdispersion). The models were implemented on software R with package TMB, based on C++ templates. Besides the full specification, models with simpler structures in the covariance matrix were considered (fixed and common variance, and ρ set to 0) and fixed dispersion. These models were applied to a dataset from the National Health and Nutrition Examination Survey, where two response variables are underdispersed and one can be considered equidispersed that were measured at 1281 subjects. The double Poisson full model specification overcame the other three competitors considering three goodness-of-fit measures: Akaike Information Criteria (AIC), Bayesian Information Criteria (BIC), and maximized log-likelihood. Consequently, it estimated parameters with smaller standard error and a greater number of significant correlation coefficients. Therefore, the proposed model can deal with multivariate count responses and measures the correlation between them taking into account the effects of the covariates.

2 January 2022 in Communications in Statistics - Simulation and Computation
Verified
Journal Article
Closed access
VRVinicius Ricardo Riffel
Vinicius Ricardo Riffel
CTCESAR TACONELI
CESAR TACONELI
PJPaulo Justiniano Ribeiro Junior
Paulo Justiniano Ribeiro Junior
Distribution selection plays a key role in parametric statistics. Choosing one from a set of candidate distributions that best fits the sample data is a recurring task in a large number of statistical analysis. The choice of a misspecified and/or poor fitted distribution can lead to unreliable results and conclusions. Although distribution selection has been widely studied regarding goodness of fit procedures, less attention has been given on how the sampling design impacts the chance of choosing the most suitable distribution. In this work we address the performance of ranked set sampling (RSS), extreme ranked set sampling (ERSS), and mixture ranked set sampling (MRSS) in distribution selection. In particular, we focus on count data, an important and very usual class of random variables. A comprehensive simulation study was carried out, accounting for five distributions: Poisson, negative binomial, geometric, zero-inflated Poisson, and zero-inflated negative binomial. Results show that RSS and extensions performed better than simple random sampling (SRS) under perfect ranking. When ranking errors are present, better efficiency depends on how accurate is the ranking criterion, and the different RSS-based designs also differ in terms of their robustness. Findings are reinforced by additional simulation based on two real data sets.