Bayesian Multilevel Models for Count Data

Olumide Sunday Adesina

doi:10.46481/jnsps.2021.168

Authors

Olumide Sunday Adesina
olumidestats@gmail.com
Department of Mathematical Sciences, Redeemer’s University, Nigeria

Keywords:

Count Data, Health, Insurance, Dispersion, Multilevel Models.

Abstract

The traditional Poisson regression model for fitting count data is considered inadequate to fit over-or under-dispersed count data and new models have been developed to make up for such inadequacies inherent in the model. In this study, Bayesian Multi-level model was proposed using the No-U-Turn Sampler (NUTS) sampler to sample from the posterior distribution. A simulation was carried out for both over-and under-dispersed data from discrete Weibull distribution. Pareto k diagnostics was implemented, and the result showed that under-dispersed and over-dispersed simulated data has all its k value to be less than 0.5, which indicate that all the observations are good. Also all WAIC were the same as LOO-IC except for Poisson in the over-dispersed simulated data. Real-life data set from National Health Insurance Scheme (NHIS) was used for further analysis. Seven multi-level models were f itted and the Geometric model outperformed other model.

Dimensions

REFERENCES

M. S. Workie & A. M. Lakew, “Bayesian count regression analysis for determinants of antenatal care service visits among pregnant women in Amhara regional state, Ethiopia” Journal of Big Data 5 (2018) https://doi.org/10.1186/s40537-018-0117-8. DOI: https://doi.org/10.1186/s40537-018-0117-8

K.H.Lee,B.ACoull,A.B.Moscicki,B.J.Paster&J.R.Starr, “Bayesian Variable Selection for Multivariate Zero-Inflated Models: Application to Microbiome Count Data”, Biostatistics 21 (2020) 499. 1. DOI: https://doi.org/10.1093/biostatistics/kxy067

F. Famoye & K.P. Singh, “Zero-inflated generalized Poisson regression model with applications to domestic violence data”, Journal of Data Science 4 (2006) 117. DOI: https://doi.org/10.6339/JDS.2006.04(1).257

O. S. Adesina, T. O. Olatayo, O. O. Agboola, & P. E. Oguntunde, “Bayesian Dirichlet process mixture prior for count data”, International Journal of Mechanical Engineering and Technology 9 (2018) 630.

D. Lambert, “Zero-inflated Poisson regression, with an application to defects in manufacturing” Technometrics 34 (1992) . DOI: https://doi.org/10.2307/1269547

F. Famoye&W.Wang,“CensoredgeneralizedPoisson regression model” Journal of Computational Statistics and Data Analysis 46 (2004) 547. DOI: https://doi.org/10.1016/j.csda.2003.08.007

W. Su & X. Wang, “Hidden Markov model in multiple testing on dependent count data”, Journal of Statistical Computation and Simulation 5 (2020) 889. DOI: https://doi.org/10.1080/00949655.2019.1710507

F. Famoye & L. Carl “Exponentiated-exponential geometric regression model”, Journal of Applied Statistics 44 (2017) 2963. DOI: https://doi.org/10.1080/02664763.2016.1267117

A. H. Lee, K. Wang, J. A Scott, K. W Yau & G. J. McLachlan, “Multilevel zero-inflated Poisson regression modelling of correlated count data with excess zeros”, Statistical Methods in Medical Research 15 (2006) 47. DOI: https://doi.org/10.1191/0962280206sm429oa

A. Moghimbeigi, M.R. Eshraghian, K. Mohammad & B. Mcardle “Multilevel zero-inflated negative binomial regression modeling for overdispersed count data with extra zeros”, Journal of Applied Statistics 35 (2008) 1193. DOI: https://doi.org/10.1080/02664760802273203

A. Almasi, M.R. Eshraghian, A. Moghimbeigic, A. Rahimib, K. Mohammad & S. Fallahigilan, “Multilevel zero-inflated generalized Poisson regression modelling for dispersed correlated count data”, Statistical Methodology 30 (2016) 1. DOI: https://doi.org/10.1016/j.stamet.2015.11.001

J. Fox & S. Weisberg, An R companion to applied regression, Second Edition, Sage (2010), ISBN-13: 978-1412975148

J. Mullahy “Specification and testing of some modified count data models”, Journal of Econometrics 33 (1986) 341. DOI: https://doi.org/10.1016/0304-4076(86)90002-3

D.Heilbron, “Generalized linear models for altered zero probabilities and overdispersion in count data” SIMS Technical Report 9, Department of Epidemiology and Biostatistics, University of California, San Francisco 9 (1989).

D.Heilbron“Zero-altered and other regression models for count data with added zeros”, Biometrical Journal 36 (1994) 531. DOI: https://doi.org/10.1002/bimj.4710360505

M. Ridout, C. G. B. Dem´etrio & J. Hinde “Models for count data with many zeros”, International Biometric Conference Cape Town (1998) 1.

P. C.Burkner. “brms: An RpackageforBayesian multilevel models using Sta”, Journal of Statistical Software (2017). DOI: https://doi.org/10.18637/jss.v080.i01

M. D. Hoffman & A. Gelman “The No-U-Turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo”, The Journal of Machine Learning Research 15 (2014) 1593.

R. Natarajan & R. E. Kass, “Reference Bayesian methods for generalized linear mixed models”, Journal of the American Statistical Association 95 (2000) 227. DOI: https://doi.org/10.1080/01621459.2000.10473916

R. E. Kass, R. Natarajan, “A default conjugate prior for variance components in generalized linear mixed models (comment on article by Browne and Draper)”, Bayesian Analysis 1 (2006) 535. DOI: https://doi.org/10.1214/06-BA117B

D. Lewandowski, D. Kurowicka & H. Joe, “Generating random correlation matrices based on vines and extended onion method”, Journal of Multivariate Analysis 100 (2009) 1989. DOI: https://doi.org/10.1016/j.jmva.2009.04.008

S. Duane, A. D. Kennedy, B. J. Pendleton & D. Roweth, “Hybrid Monte Carlo”, Physics Letters B, 195 (1987) 216. DOI: https://doi.org/10.1016/0370-2693(87)91197-X

R. M. Neal, Handbook of Markov Chain Monte Carlo, volume 2, chapter MCMCUsingHamiltonian Dynamics. CRC Press 2 (2011) 1. DOI: https://doi.org/10.1201/b10905-6

S. Watanabe, “Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory”, The Journal of Machine Learning Research 11 (2011) 3571.

A.E.Gelfand, D.K.Dey, &C.H.Hang,“Modeldeterminationusingpredictive distributions with implementation via sampling-based methods.” Technical report, DTIC Document (1992)

A. Vehtari’, A. Gelman & J. Gabry Efficient Implementation of LeaveOne-Out Cross-Validation and WAIC for Evaluating Fitted Bayesian Models”, Unpublished manuscript, (2015).

A. Gelman, J. B. Carlin, H. S. Stern & D. B. Rubin, Bayesian Data Analysis, Taylor & Francis 2 (2014). DOI: https://doi.org/10.1201/b16018

H. S. Klakattawi, V. Vinciotti & K. Yu, “A simple and adaptive dispersion regression model for count data”, Entropy 20 (2016) 1. DOI: https://doi.org/10.3390/e20020142

R Core Team. R: A language and environment for statistical computing. RFoundation for Statistical Computing, Vienna, Austria. https://www.Rproject.org (2020).

R. Vinciotti, crete “DWreg: Response”, Parametric Regression for DisR package version 2.0. https://CRAN.Rproject.org/package=DWreg (2016)

C. Kleiber & A. Zeileis, Applied Econometrics with R, New York: Springer-Verlag. ISBN 978-0-387-77316-2. URL https://CRAN.Rproject.org/package=AER (2008)

P. E. Oguntunde, A. O. Adejumo & E. A. Owoloko, “Exponential Inverse Exponential (EIE) distribution”, Asian Journal of Scientific Research, 10 (2017) 169. DOI: https://doi.org/10.3923/ajsr.2017.169.177

P. E. Oguntunde, O. A. Odetunmibi & A. O. Adejumo, “ On the exponentiated generalized Weibull distribution: a generalization of the Weibull distribution”, Indian Journal of Science and Technology 8 (2015) 1. DOI: https://doi.org/10.17485/ijst/2015/v8i35/67611