Simultaneous transformation and rounding (STAR) models for integer-valued data

DOI10.1214/20-EJS1707MaRDI QIDQ66005zbMATH OpenOpenAlexFDO

Authors Dan Kowal, Antonio Canale, Daniel R. Kowal, Antonio Canale

Publication date 1 January 2020

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1906.11653, https://projecteuclid.org/euclid.ejs/1586937696

zbMATH Keywords

predictionadditive modelsBayesian additive regression tree (BART) modelcount datanonparametric regressionsimultaneously transforming and rounding (STAR)

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Censored data models (62N01) Inference from stochastic processes and prediction (62M20) Applications of statistics to environmental and related topics (62P12)

Abstract: We propose a simple yet powerful framework for modeling integer-valued data, such as counts, scores, and rounded data. The data-generating process is defined by Simultaneously Transforming and Rounding (STAR) a continuous-valued process, which produces a flexible family of integer-valued distributions capable of modeling zero-inflation, bounded or censored data, and over- or underdispersion. The transformation is modeled as unknown for greater distributional flexibility, while the rounding operation ensures a coherent integer-valued data-generating process. An efficient MCMC algorithm is developed for posterior inference and provides a mechanism for adaptation of successful Bayesian models and algorithms for continuous data to the integer-valued data setting. Using the STAR framework, we design a new Bayesian Additive Regression Tree (BART) model for integer-valued data, which demonstrates impressive predictive distribution accuracy for both synthetic data and a large healthcare utilization dataset. For interpretable regression-based inference, we develop a STAR additive model, which offers greater flexibility and scalability than existing integer-valued models. The STAR additive model is applied to study the recent decline in Amazon river dolphins.

Recommendations

• Structured additive regression for overdispersed and zero-inflated count data
• Bayesian generalized additive models for location, scale, and shape for zero-inflated and overdispersed count data
• A flexible univariate autoregressive time-series model for dispersed count data
• scientific article; zbMATH DE number 6178864
• Integer autoregressive models with structural breaks

Cites work

• scientific article; zbMATH DE number 3759377 (Why is no real title available?)
• scientific article; zbMATH DE number 47310 (Why is no real title available?)
• scientific article; zbMATH DE number 7008320 (Why is no real title available?)
• scientific article; zbMATH DE number 6107964 (Why is no real title available?)
• scientific article; zbMATH DE number 3251902 (Why is no real title available?)
• A Useful Distribution for Fitting Discrete Data: Revival of the Conway–Maxwell–Poisson Distribution
• A flexible regression model for count data
• Applied Econometrics with R
• Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
• BART: Bayesian additive regression trees
• Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables
• Bayesian Kernel Mixtures for Counts
• Bayesian zero-inflated negative binomial regression based on Pólya-gamma mixtures
• Dynamic Bayesian influenza forecasting in the United States with hierarchical discrepancy (with discussion)
• Fast sampling of Gaussian Markov random fields
• Forecasting emergency medical service call arrival rates
• Generalized Poisson Models and their Applications in Insurance and Finance
• Generalized linear models with unknown link functions
• Integer-valued functional data analysis for measles forecasting
• Multivariate adaptive regression splines
• Multivariate output analysis for Markov chain Monte Carlo
• Negative Binomial Regression
• Nonparametric Bayes modelling of count processes
• Regression analysis of count data
• Restricted generalized poisson regression model
• Robust adaptive Metropolis algorithm with coerced acceptance rate
• Slice sampling. (With discussions and rejoinder)
• Spike-and-slab priors for function selection in structured additive regression models
• Strictly Proper Scoring Rules, Prediction, and Estimation
• The horseshoe estimator for sparse signals
• Understanding predictive information criteria for Bayesian models
• Why you cannot transform your way out of trouble for small counts

Cited in

(5)

• countSTAR
• Fast, Optimal, and Targeted Predictions Using Parameterized Decision Analysis
• Concave Likelihood-Based Regression with Finite-Support Response Variables
• Crime in Philadelphia: Bayesian Clustering with Particle Optimization
• Bayesian data synthesis and the utility-risk trade-off for mixed epidemiological data

Describes a project that uses

Uses Software

• BartPy
• COUNT
• dbarts
• splines2
• ramcmc
• rstanarm
• AER
• mcmcse
• R

This page was built for publication: Simultaneous transformation and rounding (STAR) models for integer-valued data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q66005)