
In many scientific and public health studies, response data often comprise heterogeneous outcomes, such as continuous and binary variables, along with a set of predictors collected from experiments. It is important to account for the inherent association between these heterogeneous outcomes when simultaneously addressing parameter estimation and variable selection in scenarios where the goal is to jointly model both responses within a unified modeling framework. In this paper, we present a general strategy that employs a normal linear regression for the continuous outcome and a latent variable structure to approximate logistic regression for the binary outcome. A hybrid Gibbs sampling algorithm is developed, incorporating Metropolis–Hastings steps to efficiently update model parameters, particularly those with intractable conditional distributions. The proposed sampling strategy not only improves efficiency and reliability in statistical inference but is also well-suited for high-dimensional Bayesian models and complex datasets. Numerical simulations demonstrate the effectiveness of the proposed approach, and a real-world application is provided to illustrate its practical utility.
In longitudinal health insurance studies, modeling medical expenditures and their complex covariate effects is of considerable interest. Traditional mean regression models may perform poorly in the presence of outliers or heavy-tailed responses. To address this issue, we develop a Bayesian semiparametric quantile mixed-effects model with Lasso regularization for simultaneous estimation and variable selection. Assuming the Asymmetric Laplace distribution (ALD) for the errors and using cubic splines to capture nonlinear effects, posterior inference is conducted via Gibbs sampling. The proposed method is applied to Medical Expenditure Panel Survey (MEPS) data and further assessed through simulation studies.
Traditional datasets typically represent each variable with a single value per observation. However, as data volume and complexity continue to grow, representing variables through high-level descriptors such as intervals, histograms, and probability distributions, collectively referred to as symbolic data, has become increasingly valuable. These enriched representations retain information about distributional structure and variability, thereby enhancing both analytical depth and interpretability. This paper introduces dataSDA, an R package developed to curate symbolic datasets across diverse research domains and to support their reading, writing, conversion, and summarization. Building upon the frameworks of RSDA and HistDAWass, dataSDA extends their functionality by providing unified format conversion with automatic detection, aggregation of conventional (single-valued) data into symbolic form, and functions for computing interval distances, similarity measures, and descriptive statistics. The package currently hosts 114 benchmark datasets. A subset of these is used to illustrate clustering, classification, and regression analyses, as well as exploratory data analysis and visualization with the ggInterval package. By integrating ready-to-use symbolic datasets with tools for data format transformation and descriptive statistics, dataSDA aims to serve as a comprehensive resource for symbolic data collection and analysis. The package promotes accessibility, transparency, and reproducibility in symbolic data research and is freely available on CRAN.