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

DOI10.1214/20-EJS1707zbMATH Open1439.62102arXiv1906.11653OpenAlexW3016787610MaRDI QIDQ66005FDOQ66005

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

Publication date: 1 January 2020

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

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.

Full work available at URL: https://arxiv.org/abs/1906.11653

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)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Robust adaptive Metropolis algorithm with coerced acceptance rate
• Applied Econometrics with R
• BART: Bayesian additive regression trees
• A flexible regression model for count data
• Slice sampling. (With discussions and rejoinder)
• Understanding predictive information criteria for Bayesian models
• Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
• Strictly Proper Scoring Rules, Prediction, and Estimation
• Fast sampling of Gaussian Markov random fields
• Multivariate adaptive regression splines
• Dynamic Bayesian influenza forecasting in the United States with hierarchical discrepancy (with discussion)
• Bayesian zero-inflated negative binomial regression based on Pólya-gamma mixtures
• Nonparametric Bayes modelling of count processes
• Negative Binomial Regression
• Why you cannot transform your way out of trouble for small counts
• Bayesian Kernel Mixtures for Counts
• The horseshoe estimator for sparse signals
• Restricted generalized poisson regression model
• Generalized linear models with unknown link functions
• Generalized Poisson Models and their Applications in Insurance and Finance
• Spike-and-Slab Priors for Function Selection in Structured Additive Regression Models
• Multivariate output analysis for Markov chain Monte Carlo
• Regression Analysis of Count Data
• Integer‐valued functional data analysis for measles forecasting
• Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables
• A Useful Distribution for Fitting Discrete Data: Revival of the Conway–Maxwell–Poisson Distribution
• Forecasting emergency medical service call arrival rates

Cited In (5)

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

Uses Software

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

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)