×

zbMATH — the first resource for mathematics

Thin plate regression splines. (English) Zbl 1063.62059
Summary: I discuss the production of low rank smoothers for \(d\geq 1\) dimensional data, which can be fitted by regression or penalized regression methods. The smoothers are constructed by a simple transformation and truncation of the basis that arises from the solution of the thin plate spline smoothing problem and are optimal in the sense that the truncation is designed to result in the minimum possible perturbation of the thin plate spline smoothing problem given the dimension of the basis used to construct the smoother. By making use of Lanczos iteration the basis change and truncation are computationally efficient.
The smoothers allow the use of approximate thin plate spline models with large data sets, avoid the problems that are associated with ‘knot placement’ that usually complicate modelling with regression splines or penalized regression splines, provide a sensible way of modelling interaction terms in generalized additive models, provide low rank approximations to generalized smoothing spline models, appropriate for use with large data sets, provide a means for incorporating smooth functions of more than one variable into nonlinear models and improve the computational efficiency of penalized likelihood models incorporating thin plate splines. Given that the approach produces spline-like models with a sparse basis, it also provides a natural way of incorporating unpenalized spline-like terms in linear and generalized linear models, and these can be treated just like any other model terms from the point of view of model selection, inference and diagnostics.

MSC:
62G08 Nonparametric regression and quantile regression
62J12 Generalized linear models (logistic models)
Software:
gss
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Akaike H., Proc. 2nd Int. Symp. Information Theory pp 267– (1973)
[2] DOI: 10.1139/cjfas-54-12-2727
[3] Craven P., Numer. Math. 31 pp 377– (1979)
[4] Demmel J., Applied Numerical Linear Algebra (1997)
[5] Duchon J., Construction Theory of Functions of Several Variables (1977) · Zbl 0354.28004
[6] DOI: 10.1214/ss/1038425655 · Zbl 0955.62562
[7] Gill P. E., Practical Optimization (1981) · Zbl 0503.90062
[8] Green P. J., Nonparametric Regression and Generalized Linear Models (1994) · Zbl 0832.62032
[9] Gu C., Smoothing Spline ANOVA Models (2002) · Zbl 1051.62034
[10] Gu C., SIAM J. Sci. Statist. Comput. 12 pp 383– (1991)
[11] J. R. Statist. Soc. 55 pp 353– (1993)
[12] Hastie T. J., Generalized Additive Models (1990) · Zbl 0747.62061
[13] Hutchinson M. F., Numer. Math. 47 pp 99– (1985)
[14] Parker R. L., J. R. Statist. Soc. 47 pp 40– (1985)
[15] Silverman B. W., J. R. Statist. Soc. 47 pp 1– (1985)
[16] Smola A. J., Proc. 17th Int. Conf. Machine Learning (2000)
[17] Wahba G., Approximation Theory III pp 905– (1980)
[18] J. R. Statist. Soc. 45 pp 133– (1983)
[19] -(1990 ) Spline models for observational data.CBMS-NSF Regl, Conf. Ser. Appl. Math. , 59 .
[20] DOI: 10.1214/aos/1034713638 · Zbl 0854.62042
[21] Watkins D. S., Fundamentals of Matrix Computation (1991) · Zbl 0746.65022
[22] Williams C. K. I., Advances in Neural Information Processing Systems (2001)
[23] DOI: 10.1111/1467-9868.00240 · Zbl 04558581
[24] DOI: 10.1198/016214501750332820 · Zbl 1014.62049
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.