×

Some asymptotic results on generalized penalized spline smoothing. (English) Zbl 1248.62055

Summary: The paper discusses asymptotic properties of penalized spline smoothing if the spline basis increases with the sample size. The proof is provided in a generalized smoothing model allowing for non-normal responses. The results are extended in two ways. First, assuming the spline coefficients to be a priori normally distributed links the smoothing framework to generalized linear mixed models. We consider the asymptotic rates such that the Laplace approximation is justified and the resulting fits in the mixed model correspond to the penalized spline estimates. Secondly, we make use of a fully Bayesian viewpoint by imposing an a priori distribution on all parameters and coefficients. We argue that with the postulated rates at which the spline basis dimension increases with the sample size the posterior distribution of the spline coefficients is approximately normal. The validity of this result is investigated in finite samples by comparing Markov chain Monte Carlo results with their asymptotic approximation in a simulation study.

MSC:

62G08 Nonparametric regression and quantile regression
62J12 Generalized linear models (logistic models)
62G15 Nonparametric tolerance and confidence regions
62G20 Asymptotic properties of nonparametric inference
65C40 Numerical analysis or methods applied to Markov chains
65C60 Computational problems in statistics (MSC2010)

Software:

SemiPar; BayesX; GMRFLib
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Aerts, Some theory for penalized spline additive models, J. Statist. Planng Inf. 103 pp 455– (2002) · Zbl 1011.62040
[2] Barndorff-Nielsen, Asymptotic Techniques for Use in Statistics (1989) · Zbl 0672.62024 · doi:10.1007/978-1-4899-3424-6
[3] Barrow, Asymptotic properties of best L2[0,1] approximation by splines with variable knots, Q. Appl. Math. 36 pp 293– (1978) · Zbl 0395.41012
[4] Bernardo, Bayesian Theory (2005)
[5] de Boor, A Practical Guide to Splines (2001)
[6] Breslow, Approximate inference in generalized linear mixed model, J. Am. Statist. Ass. 88 pp 9– (1993) · Zbl 0775.62195
[7] Breslow, Bias correction in generalised linear mixed models with a single component of dispersion, Biometrika 82 pp 81– (1995) · Zbl 0823.62059
[8] Brezger, Bayes X: analysing Bayesian structured additive regression models, J. Statist. Softwr. 14 pp 11– (2005) · Zbl 1088.62038
[9] Brezger, Generalized structured additive regression based on Bayesian P-splines, Computnl Statist. Data Anal. 50 pp 967– (2006) · Zbl 1431.62308
[10] Cardot, Local roughness penalties for regression splines, Computnl Statist. 17 pp 89– (2002) · Zbl 1010.62031
[11] Claeskens (2008)
[12] Demko, Inverses of band matrices and local convergence of spline projections, SIAM J. Numer. Anal. 14 pp 616– (1977) · Zbl 0367.65024
[13] Eilers, Flexible smoothing with B-splines and penalties, Statist. Sci. 11 pp 89– (1996) · Zbl 0955.62562
[14] Fahrmeir, L. and Kneib, T. (2008) Propriety of posteriors in structured additive regression models: theory and empirical evidence. J. Statist. Planng Inf., to be published. · Zbl 1156.62029
[15] Fahrmeir, Penalized structured additive regression for space-time data: a Bayesian perspective, Statist. Sin. 14 pp 715– (2004) · Zbl 1073.62025
[16] Hämmerlin, Numerische Mathematik (1992) · doi:10.1007/978-3-662-00173-8
[17] Kauermann, Penalised spline fitting in multivariable survival models with varying coefficients, Computnl Statist. Data Anal. 49 pp 169– (2005)
[18] Kneib, T. (2006) Mixed model based inference in structured additive regression. PhD Thesis. Ludwig-Maximilians-Universität München, Munich. · Zbl 1157.62506
[19] Lang, Bayesian P-splines, J. Computnl Graph. Statist. 13 pp 183– (2004)
[20] Li, On the asymptotics of penalized splines, Biometrika 95 pp 415– (2008) · Zbl 1437.62540
[21] Lin, X. and Zhang, D. (1999) Inference in generalized additive mixed models by using smoothing splines. J. R. Statist. Soc. B, 61, 381-400. · Zbl 0915.62062
[22] McCullagh, Tensor Methods in Statistics (1987)
[23] McCulloch, Generalized, Linear and Mixed Models (2001)
[24] O’Hagan, Kendall’s Advanced Theory of Statistics, vol. 2B, Bayesian Inference (2004)
[25] Opsomer, Theory for penalised spline regression, Biometrika 92 pp 105– (2005) · Zbl 1068.62045
[26] O’Sullivan, A statistical perspective on ill-posed inverse problems (with discussion), Statist. Sci. 1 pp 502– (1986)
[27] Rue, Gaussian Markov Random Fields; Theory and Applications (2005) · Zbl 1093.60003
[28] Rue, Approximate inference for hierarchical Gaussian Markov random fields models (2005) · doi:10.1201/9780203492024
[29] Ruppert, Selecting the number of knots for penalized splines, J. Computnl Graph. Statist. 11 pp 735– (2002)
[30] Ruppert, Semiparametric Regression (2003) · Zbl 1038.62042 · doi:10.1017/CBO9780511755453
[31] Severini, Likelihood Methods in Statistics (2000)
[32] Shun, Laplace approximation of high dimensional integrals, J. R. Statist. Soc. B 57 pp 749– (1995) · Zbl 0826.41026
[33] Wand, On the optimal amount of smoothing in penalised spline regression, Biometrika 86 pp 936– (1999) · Zbl 0943.62034
[34] Wand, Smoothing and mixed models, Computnl Statist. 18 pp 223– (2003)
[35] Wang, Mixed effects smoothing spline analysis of variance, J. R. Statist. Soc. B 60 pp 159– (1998) · Zbl 0909.62034
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.