arXiv subjects

Joaquín Martínez-Minaya

Publications and source records attributed to Joaquín Martínez-Minaya.

5 recordsLinked to original sources

arXiv2026

Soil Texture Prediction with Bayesian Generalized Additive Models for Spatial Compositional Data

Compositional data (CoDa) play an important role in many fields such as ecology, geology, and biology. The most widely used modeling approaches are based on the Dirichlet and the logistic-normal formulation under Aitchison geometry. Recent developments in simplex geometry allow regression models to be expressed in terms of coordinates and their coefficients to be estimated. Once the model is projected into real space, a multivariate Gaussian regression can be employed. However, most existing methods focus on linear models, and there is a lack of flexible alternatives such as additive or spatial models, especially within a Bayesian framework and with practical implementation details. In this work, we present a geoadditive regression model for CoDa from a Bayesian perspective using the \texttt{brms} package in R. The model applies the isometric log-ratio (ilr) transformation and penalized splines to incorporate nonlinear effects. We also propose two new Bayesian goodness-of-fit measures for CoDa regression: BR-CoDa-$R^2$ and BM-CoDa-$R^2$, extending the Bayesian $R^2$ to the compositional setting. {These measures, alongside WAIC, support the descriptive assessment of explained variability and complement model evaluation.} The methodology is validated through simulation studies and applied to predict soil texture composition in the Basque Country. Results demonstrate good performance, interpretable spatial patterns, and reliable quantification of explained variability in compositional outcomes.

stat.ME↗

arXiv2025

A Bayesian survival model induced by hurdle zero-modified power series discrete frailty with dispersion: an application in lung cancer

Frailty survival models are widely used to capture unobserved heterogeneity among individuals in clinical and epidemiological research. This paper introduces a Bayesian survival model that features discrete frailty induced by the hurdle zero-modified power series (HZMPS) distribution. A key characteristic of HZMPS is the inclusion of a dispersion parameter, enhancing flexibility in capturing diverse heterogeneity patterns. Furthermore, this frailty specification allows the model to distinguish individuals with higher susceptibility to the event of interest from those potentially cured or no longer at risk. We employ a Bayesian framework for parameter estimation, enabling the incorporation of prior information and robust inference, even with limited data. A simulation study is performed to explore the limits of the model. Our proposal is also applied to a lung cancer study, in which patient variability plays a crucial role in disease progression and treatment response. The findings of this study highlight the importance of more flexible frailty models in survival data analysis and emphasize the potential of the Bayesian approach to modeling heterogeneity in biomedical studies.

stat.ME↗

arXiv2024

Bayesian feedback in the framework of ecological sciences

In ecology we may find scenarios where the same phenomenon (species occurrence, species abundance, etc.) is observed using two different types of samplers. For instance, species data can be collected from scientific sampling with a completely random sample pattern, but also from opportunistic sampling (e.g., whale or bird watching fishery commercial vessels), in which observers tend to look for a specific species in areas where they expect to find Species Distribution Models (SDMs) are a widely used tool for analyzing this kind of ecological data. Specifically, we have two models available for the above data: a geostatistical model (GM) for the data coming from a complete random sampler and a preferential model (PM) for data from opportunistic sampling. Integration of information coming from different sources can be handled via expert elicitation and integrated models. We focus here in a sequential Bayesian procedure to connect two models through the update of prior distributions. Implementation of the Bayesian paradigm is done through the integrated nested Laplace approximation (INLA) methodology, a good option to make inference and prediction in spatial models with high performance and low computational costs. This sequential approach has been evaluated by simulating several scenarios and comparing the results of sharing information from one model to another using different criteria. The procedure has also been exemplified with a real dataset. Our main results imply that, in general, it is better to share information from the independent (completely random) to the preferential model than the alternative way. However, it depends on different factors such as the spatial range or the spatial arrangement of sampling locations.

stat.AP↗

arXiv2023

A flexible Bayesian tool for CoDa mixed models: logistic-normal distribution with Dirichlet covariance

Compositional Data Analysis (CoDa) has gained popularity in recent years. This type of data consists of values from disjoint categories that sum up to a constant. Both Dirichlet regression and logistic-normal regression have become popular as CoDa analysis methods. However, fitting this kind of multivariate models presents challenges, especially when structured random effects are included in the model, such as temporal or spatial effects. To overcome these challenges, we propose the logistic-normal Dirichlet Model (LNDM). We seamlessly incorporate this approach into the R-INLA package, facilitating model fitting and model prediction within the framework of Latent Gaussian Models (LGMs). Moreover, we explore metrics like Deviance Information Criteria (DIC), Watanabe Akaike information criterion (WAIC), and cross-validation measure conditional predictive ordinate (CPO) for model selection in R-INLA for CoDa. Illustrating LNDM through a simple simulated example and with an ecological case study on Arabidopsis thaliana in the Iberian Peninsula, we underscore its potential as an effective tool for managing CoDa and large CoDa databases.

stat.ME↗

arXiv2022

The Integrated Nested Laplace Approximation for fitting Dirichlet regression models

This paper introduces a Laplace approximation to Bayesian inference in Dirichlet regression models, which can be used to analyze a set of variables on a simplex exhibiting skewness and heteroscedasticity, without having to transform the data. These data, which mainly consist of proportions or percentages of disjoint categories, are widely known as compositional data and are common in areas such as ecology, geology, and psychology. We provide both the theoretical foundations and a description of how Laplace approximation can be implemented in the case of Dirichlet regression. The paper also introduces the package dirinla in the R-language that extends the R-INLA package, which can not deal directly with Dirichlet likelihoods. Simulation studies are presented to validate the good behaviour of the proposed method, while a real data case-study is used to show how this approach can be applied.

stat.CO↗