×

Flexible objective Bayesian linear regression with applications in survival analysis. (English) Zbl 1516.62577

Summary: We study objective Bayesian inference for linear regression models with residual errors distributed according to the class of two-piece scale mixtures of normal distributions. These models allow for capturing departures from the usual assumption of normality of the errors in terms of heavy tails, asymmetry, and certain types of heteroscedasticity. We propose a general non-informative, scale-invariant, prior structure and provide sufficient conditions for the propriety of the posterior distribution of the model parameters, which cover cases when the response variables are censored. These results allow us to apply the proposed models in the context of survival analysis. This paper represents an extension to the Bayesian framework of the models proposed in [the first author and Y. Hong, “Survival and lifetime data analysis with a flexible class of distributions”, ibid. 43, No. 10, 1794–1813 (2016; doi:10.1080/02664763.2015.1120710)]. We present a simulation study that shows good frequentist properties of the posterior credible intervals as well as point estimators associated to the proposed priors. We illustrate the performance of these models with real data in the context of survival analysis of cancer patients.

MSC:

62-XX Statistics

Software:

Rtwalk; t-walk
PDFBibTeX XMLCite
Full Text: DOI arXiv Link

References:

[1] R.B. Arellano-Valle, L.M. Castro, M.G. Genton, and H.W. Gómez, Bayesian inference for shape mixtures of skewed distributions, with application to regression analysis, Bayesian Anal. 3 (2008), pp. 513-540. doi: 10.1214/08-BA320 · Zbl 1330.62242 · doi:10.1214/08-BA320
[2] R.B. Arellano-Valle, H.W. Gómez, and F.A. Quintana, Statistical inference for a general class of asymmetric distributions, J. Statist. Plann. Inference 128 (2005), pp. 427-443. doi: 10.1016/j.jspi.2003.11.014 · Zbl 1095.62015 · doi:10.1016/j.jspi.2003.11.014
[3] A. Azzalini, A class of distributions which includes the normal ones, Scand. J. Statist. 12 (1985), pp. 171-178. · Zbl 0581.62014
[4] A. Azzalini and M.G. Genton, Robust likelihood methods based on the skew-\(t\) and related distributions, Int. Stat. Rev. 76 (2008), pp. 106-129. doi: 10.1111/j.1751-5823.2007.00016.x · Zbl 1206.62102 · doi:10.1111/j.1751-5823.2007.00016.x
[5] J.A. Christen and C. Fox, A general purpose sampling algorithm for continuous distributions (the t-walk), Bayesian Anal. 5 (2010), pp. 263-282. doi: 10.1214/10-BA60 · Zbl 1330.62007 · doi:10.1214/10-BA60
[6] G.B. Dantzig and M.N Thapa, Linear Programming 1: Introduction, Springer-Verlag, New York, 1997. · Zbl 0883.90090
[7] C. Fernández and M.F.J. Steel, On Bayesian modeling of fat tails and skewness, J. Amer. Statist. Assoc. 93 (1998), pp. 359-371. · Zbl 0910.62024
[8] C. Fernández and M.F.J. Steel, Bayesian regression analysis with scale mixtures of normals, Econometric Theory 16 (2000), pp. 80-101. doi: 10.1017/S0266466600161043 · Zbl 0945.62031 · doi:10.1017/S0266466600161043
[9] M.A. Juárez and M.F.J. Steel, Model-based clustering of non-Gaussian panel data based on skew-\(t\) distributions, J. Bus. Econom. Statist. 28 (2010), pp. 52-66. doi: 10.1198/jbes.2009.07145 · Zbl 1198.62097
[10] J. Lin, D. Sinha, S. Lipsitz, and A. Polpo, Semiparametric Bayesian survival analysis using models with log-linear median, Biometrics 68 (2012), pp. 1136-1145. doi: 10.1111/j.1541-0420.2012.01782.x · Zbl 1259.62020 · doi:10.1111/j.1541-0420.2012.01782.x
[11] C.L. Loprinzi, J.A. Laurie, H.S. Wieand, J.E. Krook, P.J. Novotny, J.W. Kugler, J. Bartel, M. Law, M. Bateman, N.E. Klatt, et al., Prospective evaluation of prognostic variables from patient-completed questionnaires, J. Clin. Oncol. 12 (1994), pp. 601-607. · doi:10.1200/JCO.1994.12.3.601
[12] G.S. Mudholkar and A.D. Hutson, The epsilon-skew-normal distribution for analyzing near-normal data, J. Statist. Plann. Inference 83 (2000), pp. 291-309. doi: 10.1016/S0378-3758(99)00096-8 · Zbl 0943.62012 · doi:10.1016/S0378-3758(99)00096-8
[13] B.J. Reich and L.B. Smith, Bayesian quantile regression for censored data, Biometrics 69 (2013), pp. 651-660. doi: 10.1111/biom.12053 · Zbl 1418.62170 · doi:10.1111/biom.12053
[14] V. Roy and D.K. Dey, Propriety of posterior distributions arising in categorical and survival models under generalized extreme value distribution, Statist. Sinica 24 (2014), pp. 699-722. · Zbl 1285.62055
[15] F.J. Rubio and M.G. Genton, Bayesian linear regression with skew-symmetric error distributions with applications to survival analysis. Stat. Med. (2016), in press.
[16] F.J. Rubio and Y. Hong, Survival and lifetime data analysis with a flexible class of distributions, J. Appl. Stat. 43 (2016), pp. 1794-1813, http://www.tandfonline.com/doi/full/10.1080/02664763.2015.1120710http://www.tandfonline.com/doi/full/10.1080/02664763.2015.1120710. · Zbl 1514.62834
[17] F.J. Rubio and M.F.J. Steel, Inference in two-piece location-scale models with Jeffreys priors (with discussion), Bayesian Anal. 9 (2014), pp. 1-22. doi: 10.1214/13-BA849 · Zbl 1327.62157 · doi:10.1214/13-BA849
[18] S.J. Sahu, D.K. Dey, and M.D. Branco, A new class of multivariate skew distributions with applications to Bayesian regression models, Canad. J. Statist. 2 (2003), pp. 129-150. doi: 10.2307/3316064 · Zbl 1039.62047 · doi:10.2307/3316064
[19] D. Sinha, M.H. Chen and S.K. Ghosh, Bayesian analysis and model selection for interval-censored survival data, Biometrics 55 (1999), pp. 585-590. doi: 10.1111/j.0006-341X.1999.00585.x · Zbl 1059.62715 · doi:10.1111/j.0006-341X.1999.00585.x
[20] C. Vallejos and M.F.J. Steel, Objective Bayesian survival analysis using shape mixtures of log-normal distributions, J. Amer. Statist.Assoc. 110 (2015), pp. 697-710. doi: 10.1080/01621459.2014.923316 · Zbl 1373.62506
[21] M. West, Outlier models and prior distributions in Bayesian linear regression, J. R. Stat. Soc. B 46 (1984), pp. 431-439. · Zbl 0567.62022
[22] N. Wichitaksorn, S.T.B. Choy, and R. Gerlach, A generalized class of skew distributions and associated robust quantile regression models, Canad. J. Statist. 42 (2014), pp. 579-596. doi: 10.1002/cjs.11228 · Zbl 1329.62086 · doi:10.1002/cjs.11228
[23] N. Wichitaksorn, J.J.J. Wang, S.T.B. Choy, and R. Gerlach, Analyzing return asymmetry and quantiles through stochastic volatility models using asymmetric Laplace error via uniform scale mixtures, Appl. Stoch. Models Bus. Ind. 31 (2015), pp. 584-608. doi: 10.1002/asmb.2062 · doi:10.1002/asmb.2062
[24] Z. Ying, S.H. Jung, and L.J. Wei, Survival analysis with median regression models, J. Amer. Statist. Assoc. 90 (1995), pp. 178-184. doi: 10.1080/01621459.1995.10476500 · Zbl 0818.62103
[25] K. Yu and R.A. Moyeed, Bayesian quantile regression, Statist. Probab. Lett. 54 (2001), pp. 437-447. doi: 10.1016/S0167-7152(01)00124-9 · Zbl 0983.62017 · doi:10.1016/S0167-7152(01)00124-9
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.