A Bayesian Semi-parametric Quantile Regression Approach for Joint Modeling of Longitudinal Ordinal and Continuous Responses
Abstract
Quantile regression (QR) models are one of the methods for longitudinal data analysis. When responses seemto be skew and asymmetric due to outliers and heavy-tails, QR models may work suitably. This paper developes the semi-parametric quantile regression model for analyzing longitudinal continuous and ordinal mixed responses. The latent variable model and some threshold parameters are used to perform the quantile regression model’s ordinal part. The error of the latent variable model has Asymmetric Laplace (AL) distribution. The error term’s distribution is assumed to be AL distribution to model the continuous responses. The correlations of longitudinal responses belong to the same individual and those of mixed continuous and ordinal responses are considered using a random-effects approach. The regression spline is used to approximate the non-parametric part of the model. The parameter estimation procedure is performed under aBayesian paradigm using the Gibbs sampling method. A simulation study is performed to demonstrate the proposed model’s performance where the relative biases, standard errors, and root of MSEs of estimated parameters are decreased in the semi- parametric QR joint model when the number of subjects is increased. In our application, it was found that the mother’s age and her child’s age have significant effects on reading ability, and antisocial behavior depends on the child’s gender.References
A. Beck, and M. Teboulle, Prior elicitation for mixed quantile regression with an allometric model, Environmetrics, vol. 22, no. 7, pp. 911–920, 2011.
Bahrami Samani, E., Ganjali, M., and Khodaddadi, A latent variable model for mixed continuous and ordinal responses, Journal of Statistical Theory and Applications, vol. 7, no. 3, pp. 337–349, 2008.
Bahrami Samani, E., and Ganjali, M., A multivariate latent variable model for mixed continuous and ordinal responses, World Applied Sciences Journal, vol. 3, no. 2, pp. 294–294, 2008.
Bahrami Samani, E., and Ganjali, M., Analyzing mixed correlated continuous and ordinal responses; A latent variable approach, Journal of Applied Statistical Science, vol. 17, no. 4, pp. 1–12, 2009.
Bahrami Samani, E., Ganjali, M., and Eftekhari, S., A latent variable model for mixed continuous and ordinal responses with non-ignorable missing responses: Assessing the local influence via covariance structure, Sankhya B, vol. 72, no. 1, pp. 38–57, 2010.
Bahrami Samani, E.,Ganjali, M.,andAmirian,Y., Likelihood estimation for longitudinal zero-inated power series regression models, Journal of Applied Statistics, vol. 39, no. 9, pp. 1965–1974, 2010.
Cai, Z. and Xiao, Z., Semiparametric quantile regression estimation in dynamic models with partially varying coefficients, Journal of Econometrics, vol. 167, no. 2, pp. 413–425, 2015.
Chen, X., Sun, J. and Liu, L., Semiparametric partial linear quantile regression of longitudinal data with time varying coefficients and informative observation times, Statistica Sinica, vol. 25, no. 4, pp. 1437–1458, 2015.
de Boor, C., A Practical Guide to Splines, Springer-Verlag, New York, 1978.
Eubank, R.L., NonparametricRegression and Spline Smoothing, Marcel Dekker. New York, 1999.
Fan, J., Design-adaptive nonparametric regression, Journal of American Statistical Association, vol. 87, pp. 998–1004, 1992.
Fan, J., Local linear regression smoothers and their minimax efficiency, Annals of Statistics, vol. 21, pp. 196–216, 1993.
Fan, J. and Gijbels, I, Variable bandwidth and local linear regression smoothers, Annals of Statistics, vol. 20, pp. 2008–2036, 1992.
Fitzmaurice, G., Davidian, M., Verbeke, G., and Molenberghs, G., Longitudinal data analysis, Boston, CRC Press, 2008.
Fu, L. and Wang, Y.G., Quantile regression for longitudinal data with a working correlation model, Comput.Stat.Data Anal, vol. 56, pp. 2526–2538, 2012.
Ganjali, M., A model for mixed continuous and discrete responses with possibility of missing responses, Journal of Sciences Islamic Republic of Iran, vol. 14, no. 1, pp. 53–60, 2003.
Gelman, and Rubin, D. B., Inference from iterative simulation using sequences, Statistical Science, vol. 7, pp. 457–511, 1992.
Geraci, M., and Bottai, M., Quantile regression for longitudinal data using the asymmetric Laplace distribution, Biostatistics, vol. 8, no. 1, pp. 140–154, 2007.
Ghasemzadeh, S., Ganjali, M., and Baghfalaki, T., Bayesian quantile regression for joint modeling of longitudinal mixed ordinal and continuous data, Communications in Statistics-Simulation and Computation, vol. 49, no. 2, pp. 375–395, 2020.
Greene W. H., and Hensher D. A., Modeling ordered choices: A primer, Cambridge University Press, 2010.
Hao, L. and Naiman, D. Q., Quantile Regression, New York, Sage, 2007.
Heckman, J. J. D., Dummy Endogenous variable in a simultaneous Equations system, Econometrica, vol. 46, no. 6, pp. 931–959, 1978.
Hong, H.G., He, X., Prediction of functional status for the elderly based on a new ordinal regression model, Journal of American Statistical Association, vol. 105, no. 491, pp. 930–941, 2010.
Hong, H.G., Zhou, J., A multi-index model for quantile regression with ordinal data, Journal of Applied Statistics, vol. 40, no. 6, pp. 1231–1245, 2013.
Huang, Y., Quantile regression-based Bayesian semiparametric mixed-effects models for longitudinal data with non normal, missing and mismeasured covariate, Journal of Statistical Computation and Simulation, vol. 86, no. 6, pp. 1183–1202, 2016.
Jeliazkov, I., Graves, J., and Kutzbach, M., Fitting and comparison of models for multivariate ordinal outcomes, Advances in Econometrics: Bayesian Econometrics, vol. 23, pp. 115–156, 2008.
Karlsson, A., Nonlinear quantile regression estimation of longitudinal data, Communications in Statistics-Simulation and Computation, vol. 37, no. 1, pp. 114–131, 2007.
Koenker, R. and Bassett, G., Regression quantiles, Econometrica, vol. 46, pp. 33–50, 1978.
Koenker, R., Quantile regression for longitudinal data, J. Multivar. Anal, vol. 91, pp. 74–89, 2004.
Koenker, R., Quantile regression, Cambridge, Cambridge university press, 2005.
Kottas, A. and Krnjajic, M., Bayesian Semiparametric Modelling in Quantile Regression, Scandinavian Journal of Statistics, vol. 36, pp. 297–319, 2009.
Kotz, S., Kozubowski, T.J., Podgorski, K., The Laplace distribution and generalizations: a revisit with applications to communications, economics, engineering, and finance, Boston:Birkhauser, 2001.
Kozumi, H., and Kobayashi, G., Gibbs sampling methods for Bayesian quantile regression, Journal of Statistical Computation and Simulation, vol. 81, no. 11, pp. 1565–1578, 2011.
Leon, A. R., and Chough, K. C., Analysis of mixed data: Methods and applications, London: Chapman and Hall/CRC, 2013.
McCullagh, P., Regression models for ordinal data (with discussion), Journal of the Royal Statistical Society, Series B, vol. 42, no. 2, pp. 109–142, 1980.
McKelvey, R., and Zavoina, W., An IBM Fortran IV Program to perform N-Chotomus multivariate probit aAnalysis, Behavioral Science, vol. 16, no. 2, pp. 186–187, 1971.
McKelvey, R., and Zavoina, W., A statistical model for the analysis of ordered level dependent variables, Journal of Mathematical Sociology, vol. 4, pp. 103–120, 1975.
Molenberghs, G. and Verbeke, G., Models for discrete longitudinal data, New York, Springer, 2005.
Noorian, S.,Ganjali, M., and Bahrami Samani, E., A Bayesian test of homogeneity of association parameter using transition modeling of longitudinal mixed responses, Journal of Applied Statistics, vol. 43, no. 10, pp. 1850–1863, 2016.
Olkin, L.,and Tate, R.F., Multivariate correlation models with mixed discrete and continuous variables, The Annals of Mathematical Statistics, vol. 32, pp. 448–465, 1961.
R Core Team, R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria. Available at http://www.R-project.org/, 2013.
Samani, E. B., and Ganjali, M., Bayesian multivariate latent variable model for mixed correlated ordinal and continuous responses, Journal of Statistical Research, vol. 43, no. 2, 2009.
Samani, E. B., and Ganjali, M., Sensitivity analysis for non-ignorable missing responses with application to multivariate Random effect model, Metron, vol. 69, no. 3, pp. 309–322, 2011.
Samuel, M., Ryan, L., and Legler JM., Latent variable models for mixed discreteand continuous outcomes, Journal of the Royal Statistical Society,Series B: Methodological, vol. 59, pp. 667–678, 1997.
Sharifian, N., Samani, E. B., and Ganjali, M., Joint modeling for longitudinal set-inflated continuous and count responses, Communications in Statistics - Theory and Methods, vol. 50, no. 5, pp. 1134–1160, 2021. doi:10.1080/03610926.2019.1646768.
Sharifian, N., Samani, E. B., and Ganjali, M., Joint model for longitudinal mixture of normal and zero- inflated power series correlated responses, Journal of Biopharmaceutical Statistics, vol. 31, no. 2, pp. 117–140, 2021. doi:10.1080/10543406.2020.1814798.
Smith, P.L., Splines as a usefid and convenient statistical tool, American Statistician, vol. 33, pp. 57–62, 2015.
Spiegelhalter, D. J., Best, N. G., Carlin, B. P., and Van Der Linde, A., Bayesian measures of model complexity and fit, Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 64, no. 4, pp. 583–639, 2002.
Teimourian, M., Baghfalaki, T., Ganjali, M., and Berridge, D., Joint modeling of mixed skewed continuous and ordinal longitudinal responses: a Bayesian approach, Journal of Applied Statistics, vol. 42, no. 10, pp. 2233–2256, 2015.
Wahba, G., Spline Models for Observational Data, SIAM, Philadelphia. CBMS-NSF Regional Conference Series in Applied Mathematics, Vol. 59, 1990.
Wold, S., Spline functions in data analysis, Technometrics, vol. 16, pp. 1–1, 1974.
Wu, H., and Zhang, J. T., Nonparametric regression methods for longitudinal data analysis: mixed-effects modeling approaches, John Wiley and Sons, 2006.
Xue, L., and A. Qu., Variable selection in high-dimensional varying coefficient models with global optimality, Journal of Machine Learning Research, vol. 13, pp. 1973–1998, 2012.
Yu, K., Moyeed, R.A., Bayesian quantile regression, Stat. Probab.Lett, vol. 54, pp. 437–447, 2001.
Yu, K., and Zhang, J., A three-parameter asymmetric Laplace distribution and its extension, Communications in Statistics-Theory and Methods, vol. 34, no. 9-10, pp. 1867–1879, 2015.
Zhou, L. Huang and M. Li, R., Semiparametric quantile regression with high-dimensional covariates, Statistica Sinica, vol. 22, pp.1379–1401, 2012.
Zhou, L., Conditional quantile estimation with ordinal data, Ph.D. thesis, University of South Carolina, 2010.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
