Rgbp: An R package for gaussian, poisson, and binomial random effects models with frequency coverage evaluations

Hyungsuk Tak, Joseph Kelly, Carl Morris

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Rgbp is an R package that provides estimates and verifiable confidence intervals for random effects in two-level conjugate hierarchical models for overdispersed Gaussian, Poisson, and binomial data. Rgbp models aggregate data from k independent groups summarized by observed sufficient statistics for each random effect, such as sample means, possibly with covariates. Rgbp uses approximate Bayesian machinery with unique improper priors for the hyper-parameters, which leads to good repeated sampling coverage properties for random effects. A special feature of Rgbp is an option that generates synthetic data sets to check whether the interval estimates for random effects actually meet the nominal confidence levels. Additionally, Rgbp provides inference statistics for the hyper-parameters, e.g., regression coefficients.

Original languageEnglish (US)
JournalJournal of Statistical Software
Volume78
DOIs
StatePublished - Jan 1 2017

Fingerprint

Binomial Model
Random Effects Model
Random Effects
Siméon Denis Poisson
Coverage
Statistics
Hyperparameters
Evaluation
Machinery
Improper Prior
Sampling
Sufficient Statistics
Sample mean
Confidence Level
Hierarchical Model
Regression Coefficient
Synthetic Data
Estimate
Categorical or nominal
Confidence interval

All Science Journal Classification (ASJC) codes

  • Software
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{a2ffd8ba0f6b45b28dab5bfed93bb250,
title = "Rgbp: An R package for gaussian, poisson, and binomial random effects models with frequency coverage evaluations",
abstract = "Rgbp is an R package that provides estimates and verifiable confidence intervals for random effects in two-level conjugate hierarchical models for overdispersed Gaussian, Poisson, and binomial data. Rgbp models aggregate data from k independent groups summarized by observed sufficient statistics for each random effect, such as sample means, possibly with covariates. Rgbp uses approximate Bayesian machinery with unique improper priors for the hyper-parameters, which leads to good repeated sampling coverage properties for random effects. A special feature of Rgbp is an option that generates synthetic data sets to check whether the interval estimates for random effects actually meet the nominal confidence levels. Additionally, Rgbp provides inference statistics for the hyper-parameters, e.g., regression coefficients.",
author = "Hyungsuk Tak and Joseph Kelly and Carl Morris",
year = "2017",
month = "1",
day = "1",
doi = "10.18637/jss.v078.i05",
language = "English (US)",
volume = "78",
journal = "Journal of Statistical Software",
issn = "1548-7660",
publisher = "University of California at Los Angeles",

}

Rgbp : An R package for gaussian, poisson, and binomial random effects models with frequency coverage evaluations. / Tak, Hyungsuk; Kelly, Joseph; Morris, Carl.

In: Journal of Statistical Software, Vol. 78, 01.01.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Rgbp

T2 - An R package for gaussian, poisson, and binomial random effects models with frequency coverage evaluations

AU - Tak, Hyungsuk

AU - Kelly, Joseph

AU - Morris, Carl

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Rgbp is an R package that provides estimates and verifiable confidence intervals for random effects in two-level conjugate hierarchical models for overdispersed Gaussian, Poisson, and binomial data. Rgbp models aggregate data from k independent groups summarized by observed sufficient statistics for each random effect, such as sample means, possibly with covariates. Rgbp uses approximate Bayesian machinery with unique improper priors for the hyper-parameters, which leads to good repeated sampling coverage properties for random effects. A special feature of Rgbp is an option that generates synthetic data sets to check whether the interval estimates for random effects actually meet the nominal confidence levels. Additionally, Rgbp provides inference statistics for the hyper-parameters, e.g., regression coefficients.

AB - Rgbp is an R package that provides estimates and verifiable confidence intervals for random effects in two-level conjugate hierarchical models for overdispersed Gaussian, Poisson, and binomial data. Rgbp models aggregate data from k independent groups summarized by observed sufficient statistics for each random effect, such as sample means, possibly with covariates. Rgbp uses approximate Bayesian machinery with unique improper priors for the hyper-parameters, which leads to good repeated sampling coverage properties for random effects. A special feature of Rgbp is an option that generates synthetic data sets to check whether the interval estimates for random effects actually meet the nominal confidence levels. Additionally, Rgbp provides inference statistics for the hyper-parameters, e.g., regression coefficients.

UR - http://www.scopus.com/inward/record.url?scp=85020483842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020483842&partnerID=8YFLogxK

U2 - 10.18637/jss.v078.i05

DO - 10.18637/jss.v078.i05

M3 - Article

AN - SCOPUS:85020483842

VL - 78

JO - Journal of Statistical Software

JF - Journal of Statistical Software

SN - 1548-7660

ER -