COBS: qualitatively constrained smoothing via linear programming

SummaryPopular smoothing techniques generally have a difficult time accommodating qualitative constraints like monotonicity, convexity or boundary conditions on the fitted function. In this paper, we attempt to bring the problem of constrained spline smoothing to the foreground and describe the details of a constrained B-spline smoothing (COBS) algorithm that is being made available to S-plus-users. Recent work of He & Shi (1998) considered a special case and showed that the L1 projection of a smooth function into the space of B-splines provides a monotone smoother that is flexible, efficient and achieves the optimal rate of convergence. Several options and generalizations are included in COBS: it can handle small or large data sets either with user interaction or full automation. Three examples are provided to show how COBS works in a variety of real-world applications.

[1]  Christine Thomas-Agnan,et al.  A shape constrained smoother: simulation study , 1995 .

[2]  J. Friedman.  A VARIABLE SPAN SMOOTHER , 1984 .

[3]  R. Koenker,et al.  The Gaussian hare and the Laplacian tortoise: computability of squared-error versus absolute-error estimators , 1997 .

[4]  J. Ramsay.  Estimating smooth monotone functions , 1998 .

[5]  Pin T. Ng.  An algorithm for quantile smoothing splines , 1996 .

[6]  A. Raftery,et al.  Space-time modeling with long-memory dependence: assessing Ireland's wind-power resource. Technical report , 1987 .

[7]  R. Koenker,et al.  Robust Tests for Heteroscedasticity Based on Regression Quantiles , 1982 .

[8]  C. R. Deboor,et al.  A practical guide to splines , 1978 .

[9]  B. Silverman,et al.  Nonparametric Regression and Generalized Linear Models: A roughness penalty approach , 1993 .

[10]  Douglas M. Hawkins,et al.  Fitting Monotonic Polynomials to Data , 1994 .

[11]  R. Koenker,et al.  Quantile spline models for global temperature change , 1994 .

[12]  G. S. Watson,et al.  Smooth regression analysis , 1964 .

[13]  Xuming He,et al.  Monotone B-Spline Smoothing , 1998 .

[14]  W. Cleveland.  Robust Locally Weighted Regression and Smoothing Scatterplots , 1979 .

[15]  Paul Dierckx,et al.  Curve and surface fitting with splines , 1994, Monographs on numerical analysis.

[16]  Stephen Portnoy,et al.  Local asymptotics for quantile smoothing splines , 1997 .

[17]  J. Ramsay.  Monotone Regression Splines in Action , 1988 .

[18]  Pin T. Ng,et al.  Quantile smoothing splines , 1994 .

[19]  W. Härdle.  Applied Nonparametric Regression , 1991 .

[20]  D. W. Scott,et al.  The L 1 Method for Robust Nonparametric Regression , 1994 .

[21]  B. Silverman,et al.  Nonparametric regression and generalized linear models , 1994 .

[22]  B. Silverman,et al.  Nonparametric Regression and Generalized Linear Models: A roughness penalty approach , 1993 .

[23]  G. Wahba.  Spline models for observational data , 1990 .

[24]  Richard H. Bartels,et al.  Linearly Constrained Discrete I1 Problems , 1980, TOMS.

[25]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[26]  Pin T. Ng,et al.  A remark on Bartels and Conn's linearly constrained, discrete l1 problems , 1996, TOMS.

[27]  Stephen Portnoy,et al.  Bivariate quantile smoothing splines , 1998 .

[28]  J. Hansen,et al.  Global surface air temperatures: update through 1987 , 1988 .

[29]  Edward J. Wegman,et al.  Isotonic, Convex and Related Splines , 1980 .

[30]  J. Hansen,et al.  Global trends of measured surface air temperature , 1987 .

[31]  Paul H. C. Eilers,et al.  Flexible smoothing with B-splines and penalties , 1996 .

[32]  W. Härdle.  Applied Nonparametric Regression , 1992 .

[33]  P. Shi,et al.  Convergence rate of b-spline estimators of nonparametric conditional quantile functions ∗ , 1994 .

[34]  L. Schumaker.  Spline Functions: Basic Theory , 1981 .