Bayesian Group and Sparse-Group LASSO with Spike-and-Slab Priors in Quantile Mixed Models: An Application to Child Growth Data

Taweesak  Channgam; Craig  Anderson; Tereza  Neocleous

Authors

Taweesak Channgam Department of Statistics and Information Management, Faculty of Science, Maejo University, Chiang Mai, Thailand
Craig Anderson School of Mathematics and Statistics, University of Glasgow, Glasgow, UK
Tereza Neocleous School of Mathematics and Statistics, University of Glasgow, Glasgow, UK

Keywords:

Bayesian method, group LASSO, sparse group LASSO, MCMC algorithm, quantile regression, mixed models, longitudinal data, child growth

Abstract

Identifying risk factors that exhibit significant associations with child growth and development is crucial for preventing unhealthy growth and supporting children’s overall development. Given that children have a diverse range of growth patterns, it is particularly relevant to evaluate these associations across quantiles rather than simply focusing on the mean or median values. In this paper, we
develop a Bayesian variable selection method within quantile regression (QR) and quantile mixed models (QMMs). In particular, these methods are designed to analyse longitudinal data, such as child growth data. This novel methodology combines several key components, including the Bayesian sparse group LASSO method, a likelihood function based on the scale mixture representation of the asymmetric Laplace (AL) distribution. It also incorporates spike-and-slab priors for regression coefficients and utilises linear mixed models based on a decomposition for the covariance matrix of random effects. By combining these elements, our approach offers a comprehensive solution for simultaneous selection and estimation of fixed and random effects in QMMs. We assess the performance of the proposed method through simulation studies, which demonstrate its strong variable selection and predictive capabilities. Furthermore, we illustrate its practical utility by applying it to the Growing Up in Scotland (GUS) dataset, providing practical insights into its real-world applicability.

References

Alhamzawi R. Brq: Bayesian Analysis of Quantile Regression Models; 2020, https://CRAN.R-project.org/package=Brq, R package version 3.0.

Alhamzawi R, Ali HTM. The Bayesian adaptive lasso regression. Math Biosci. 2018; 303: 75–82.

Alhamzawi R, Ali HTM. Brq: an R package for Bayesian quantile regression. Metron. 2020; 78(3): 313–328.

Alhamzawi R, Yu K. Conjugate priors and variable selection for Bayesian quantile regression. Comput Stat Data Anal. 2013; 64: 209–219.

Alhamzawi R, Yu K. Bayesian Lasso-mixed quantile regression. J Stat Comput Simul. 2014; 84(4): 868–880.

Alhamzawi R. Prior elicitation and variable selection for bayesian quantile regression. PhD [dissertation] . London, Brunel University;2013.

Andrews DF, Mallows CL. Scale mixtures of normal distributions. J R Stat Soc Ser B Stat Methodol. 1974; 36(1): 99–102.

Aniley TT, Debusho LK, Nigusie ZM, Yimer WK, Yimer BB. A semi-parametric mixed models for longitudinally measured fasting blood sugar level of adult diabetic patients. BMC Med Res Methodol. 2019; 19(1): 13.

Berk LE. Child development. Boston: Pearson; 2013.

Berk LE, Meyers AB. Infants, children, and adolescents. Boston: Pearson; 2016.

Birtwistle S, Deakin E, Whitford R, Hinchliffe S, Daniels-Creasey A, Rule S, The Scottish Health Survey: 2022 edition; [ISBN: 9781835216569]. 2023. [cited 2023 Nov 29]. Available from: https://www.gov.scot/publications/scottish-health-survey-2022-volume-1-main-report/pages/2/.

Black RE, Allen LH, Bhutta ZA, et al. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet. 2008; 371(9608): 243–260.

Bradshaw P, Tipping S, Marryat L, Corbett J. Growing up in Scotland Sweep 1 – 2005: User guide. Edinburgh, UK: Scottish Centre for Social Research; 2007.

Bukatko D, Daehler MW. Child development: a thematic approach. Boston: Houghton Mifflin; 1995.

Casella G. Empirical bayes gibbs sampling. Biostatistics. 2001; 2(4): 485–500.

Chen CWS, Dunson DB, Reed C, Yu K. Bayesian variable selection in quantile regression. Stat Its Interface. 2013; 6(2): 261–274.

Chen Z, Dunson DB. Random effects selection in linear mixed models. Biometrics. 2003; 59(4): 762–769.

Cole TJ. Growth charts for both cross-sectional and longitudinal data. Stat Med. 1994; 13(23-24): 2477–2492.

Cole TJ, Donaldson MDC, Ben-Shlomo Y. SITAR—a useful instrument for growth curve analysis. Int J Epidemiol. 2010; 39(6): 1558–1566.

Denham SA. Social-emotional competence as support for school readiness: what is it and how do we assess it? EE&D. 2006; 17(1): 57–89.

Durb´an M, Harezlak J, Wand MP, Carroll RJ. Simple fitting of subject-specific curves for longitudinal data. Stat Med. 2005; 24(8): 1153–1167.

Fitzmaurice GM, Ravichandran C. A primer in longitudinal data analysis. Circ. 2008; 118(19): 2005–2010.

Gelman A. Prior distributions for variance parameters in hierarchical models (Comment on article by browne and draper). Bayesian Anal. 2006; 1(3): 515–534.

Gelman A. Bayesian data analysis. Boca Raton: CRC Press; 2014.

Gelman A, Rubin DB. Inference from iterative simulation using multiple sequences. Stat Sci. 1992; 7(4).

George EI, McCulloch RE. Approaches for bayesian variable selection. Stat Sin. 1997; 7(2): 339–

Geraci M. Additive quantile regression for clustered data with an application to children’s physical activity. J R Stat Soc Ser C Appl Stat. 2019; 68(4): 1071–1089.

Geraci M, Bottai M. Linear quantile mixed models. Stat Comput. 2014; 24(3): 461–479.

Grammer JK, Coffman JL, Ornstein PA, Morrison FJ. Change over time: conducting longitudinal studies of children’s cognitive development. J Cogn Dev. 2013; 14(4): 515–528.

Ishwaran H, Rao JS. Spike and slab variable selection: frequentist and bayesian strategies. Ann Stat.

; 33(2): 730–773.

Ji Y, Shi H. Shrinkage estimation of fixed and random effects in linear quantile mixed models. J Appl Stat. 2022; 49(14): 3693–3716.

Kinney SK, Dunson DB. Fixed and random effects selection in linear and logistic models. Biometrics. 2007; 63(3): 690–698.

Koenker R. Quantile regression. Cambridge: Cambridge University Press; 2005.

Koenker R. quantreg: Quantile Regression; 2021, https://CRAN.R-project.org/package=quantreg, R package version 5.86.

Koenker R, Machado JAF. Goodness of Fit and Related Inference Processes for Quantile Regression.

J Am Stat Assoc. 1999; 94(448): 1296–1310.

Kozumi H, Kobayashi G. Gibbs sampling methods for Bayesian quantile regression. J Stat Comput Simul. 2011; 81(11): 1565–1578.

Kyung M, Gill J, Ghosh M, Casella G. Penalized regression, standard errors, and Bayesian LASSOS. Bayesian Anal. 2010; 5(2): 369–411.

Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982; 38(4): 963.

Li Q, Lin N, Xi R. Bayesian regularized quantile regression. Bayesian Anal. 2010; 5(3): 533–556.

Mitchell TJ, Beauchamp JJ. Bayesian variable selection in linear regression. J Am Stat Assoc. 1988; 83(404): 1023–1032.

Muggeo VM, Atkins DC, Gallop RJ, Dimidjian S. Segmented mixed models with random change points: a maximum likelihood approach with application to treatment for depression study. Stat Model. 2014; 14(4): 293–313.

Park T, Casella G. The Bayesian lasso. J Am Stat Assoc. 2008; 103(482): 681–686.

R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria; 2024, https://www.R-project.org/.

Royal College of Physicians and Surgeons of Glasgow, Obesity Action Scotland | Providing leadership and advocacy on preventing & reducing obesity & overweight in Scotland | 5 years on, are we on track to halve childhood obesity in Scotland by 2030?; 2023. [cited 2023 Nov 29]. Available from: https://www.obesityactionscotland.org/blogs/5-years-on-are-we-on-track-to-halve-childhood-obesity-in-scotland-by-2030.

Ruppert D, Wand MP, Carroll RJ. Semiparametric regression. Cambridge: Cambridge University Press; 2003.

Sabates R, Dex S. The impact of multiple risk factors on young children’s cognitive and behavioural development. Child Soc. 2015; 29(2): 95–108.

Sherwood B, Maidman A, Li S. rqPen: Penalized Quantile Regression; 2023, https://CRAN.R-project.org/package=rqPen, R package version 3.1.3.

Shonkoff JP, Phillips DA, National Research Council (U S ), editors. From neurons to neighborhoods: the science of early child development. Washington (DC): National Academy Press; 2000.

Simon N, Friedman J, Hastie T, Tibshirani R. A sparse-group lasso. J Comput Graph Stat. 2013; 22(2): 231–245.

Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005; 83(3): 457–502.

Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc Ser B Stat Methodol. 1996; 58(1): 267–288.

Tsionas EG. Bayesian quantile inference. J Stat Comput Simul. 2003; 73(9): 659–674.

Weiss ST, Ware JH. Overview of issues in the longitudinal analysis of respiratory data. Am J Respir Crit Care Med. 1996; 154(6 pt 2): S208–S211.

Wood SN. Generalized additive models: an introduction with R. Boca Raton: Chapman & Hall/CRC; 2006.

Xu X, Ghosh M. Bayesian variable selection and estimation for group lasso. Bayesian Anal. 2015; 10(4): 909–936.

Yu K, Moyeed RA. Bayesian quantile regression. Stat Probab Lett. 2001; 54(4): 437–447.

Yuan M, Lin Y. Model selection and estimation in regression with grouped variables. J R Stat Soc Ser B Stat Methodol. 2006; 68(1): 49–67.

Yue YR, Rue H. Bayesian inference for additive mixed quantile regression models. Comput Stat Data Anal. 2011; 55(1): 84–96.

Zhang L, Baladandayuthapani V, Mallick BK, et al. Bayesian hierarchical structured variable selection methods with application to molecular inversion probe studies in breast cancer. J R Stat Soc Ser C Appl Stat. 2014; 63(4): 595–620.