Adaptive piecewise polynomial estimation via trend filtering

We study trend filtering, a recently proposed tool of Kim et al. [SIAM Rev. 51 (2009) 339-360] for nonparametric regression. The trend filtering estimate is defined as the minimizer of a penalized least squares criterion, in which the penalty term sums the absolute $k$th order discrete derivatives over the input points. Perhaps not surprisingly, trend filtering estimates appear to have the structure of $k$th degree spline functions, with adaptively chosen knot points (we say ``appear'' here as trend filtering estimates are not really functions over continuous domains, and are only defined over the discrete set of inputs). This brings to mind comparisons to other nonparametric regression tools that also produce adaptive splines; in particular, we compare trend filtering to smoothing splines, which penalize the sum of squared derivatives across input points, and to locally adaptive regression splines [Ann. Statist. 25 (1997) 387-413], which penalize the total variation of the $k$th derivative. Empirically, we discover that trend filtering estimates adapt to the local level of smoothness much better than smoothing splines, and further, they exhibit a remarkable similarity to locally adaptive regression splines. We also provide theoretical support for these empirical findings; most notably, we prove that (with the right choice of tuning parameter) the trend filtering estimate converges to the true underlying function at the minimax rate for functions whose $k$th derivative is of bounded variation. This is done via an asymptotic pairing of trend filtering and locally adaptive regression splines, which have already been shown to converge at the minimax rate [Ann. Statist. 25 (1997) 387-413]. At the core of this argument is a new result tying together the fitted values of two lasso problems that share the same outcome vector, but have different predictor matrices.

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

[2]  Carl de Boor,et al.  A Practical Guide to Splines , 1978, Applied Mathematical Sciences.

[3]  M. Nussbaum Spline Smoothing in Regression Models and Asymptotic Efficiency in $L_2$ , 1985 .

[4]  B. Yandell,et al.  Semi-Parametric Generalized Linear Models. , 1985 .

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

[6]  Grace Wahba,et al.  Spline Models for Observational Data , 1990 .

[7]  R. Tibshirani,et al.  Generalized Additive Models , 1991 .

[8]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[9]  I. Daubechies,et al.  Wavelets on the Interval and Fast Wavelet Transforms , 1993 .

[10]  George G. Lorentz,et al.  Constructive Approximation , 1993, Grundlehren der mathematischen Wissenschaften.

[11]  D. Cox Nonparametric Regression and Generalized Linear Models: A roughness penalty approach , 1993 .

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

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

[14]  I. Johnstone,et al.  Adapting to Unknown Smoothness via Wavelet Shrinkage , 1995 .

[15]  S. Geer,et al.  Locally adaptive regression splines , 1997 .

[16]  Michael Unser,et al.  Ten good reasons for using spline wavelets , 1997, Optics & Photonics.

[17]  I. Johnstone,et al.  Minimax estimation via wavelet shrinkage , 1998 .

[18]  S. Mallat A wavelet tour of signal processing , 1998 .

[19]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[20]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[21]  R. Tibshirani,et al.  Sparsity and smoothness via the fused lasso , 2005 .

[22]  Michael Elad,et al.  Analysis versus synthesis in signal priors , 2006, 2006 14th European Signal Processing Conference.

[23]  S. Rosset,et al.  Piecewise linear regularized solution paths , 2007, 0708.2197.

[24]  Stephen P. Boyd,et al.  1 Trend Filtering , 2009, SIAM Rev..

[25]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[26]  GergHo Nemes,et al.  On the Coefficients of the Asymptotic Expansion of n , 2010, 1003.2907.

[27]  R. Tibshirani,et al.  The solution path of the generalized lasso , 2010, 1005.1971.

[28]  Adam D. Myers,et al.  THE BOSS Lyα FOREST SAMPLE FROM SDSS DATA RELEASE 9 , 2012, 1211.5146.

[29]  R. Tibshirani,et al.  Degrees of freedom in lasso problems , 2011, 1111.0653.

[30]  Jinglai Shen,et al.  Smoothing splines with varying smoothing parameter , 2013, 1306.1868.

[31]  Ryan J. Tibshirani,et al.  Efficient Implementations of the Generalized Lasso Dual Path Algorithm , 2014, ArXiv.