Introduction to Smoothing Splines.ppt

1、Introduction to Smoothing Splines,Tongtong Wu Feb 29, 2004,Outline,Introduction Linear and polynomial regression, and interpolation Roughness penalties Interpolating and Smoothing splines Cubic splines Interpolating splines Smoothing splines Natural cubic splines Choosing the smoothing parameter Ava

2、ilable software,Key Words,roughness penalty penalized sum of squares natural cubic splines,Motivation,Motivation,Motivation,Motivation,Spline(y18),Introduction,Linear and polynomial regression : Global influence Increasing of polynomial degrees happens in discrete steps and can not be controlled con

3、tinuously Interpolation Unsatisfactory as explanations of the given data,Roughness penalty approach,A method for relaxing the model assumptions in classical linear regression along lines a little different from polynomial regression.,Roughness penalty approach,Aims of curving fitting A good fit to t

4、he data To obtain a curve estimate that does not display too much rapid fluctuation Basic idea: making a necessary compromise between the two rather different aims in curve estimation,Roughness penalty approach,Quantifying the roughness of a curve An intuitive way:(g: a twice-differentiable curve) M

5、otivation from a formalization of a mechanical device: if a thin piece of flexible wood, called a spline, is bent to the shape of the graph g, then the leading term in the strain energy is proportional to,Roughness penalty approach,Penalized sum of squaresg: any twice-differentiable function on a,b

6、: smoothing parameter (rate of exchange between residual error and local variation) Penalized least squares estimator,Roughness penalty approach,Curve for a large value of,Roughness penalty approach,Curve for a small value of,Interpolating and Smoothing Splines,Cubic splines Interpolating splines Sm

7、oothing splines Choosing the smoothing parameter,Cubic Splines,Given at1t2tnb, a function g is a cubic spline if On each interval (a,t1), (t1,t2), , (tn,b), g is a cubic polynomial The polynomial pieces fit together at points ti (called knots) s.t. g itself and its first and second derivatives are c

8、ontinuous at each ti, and hence on the whole a,b,Cubic Splines,How to specify a cubic splineNatural cubic spline (NCS) if its second and third derivatives are zero at a and b, which implies d0=c0=dn=cn=0, so that g is linear on the two extreme intervals a,t1 and tn,b.,Natural Cubic Splines,Value-sec

9、ond derivative representation We can specify a NCS by giving its value and second derivative at each knot ti. Definewhich specify the curve g completely. However, not all possible vectors represent a natural spline!,Natural Cubic Splines,Value-second derivative representation Theorem 2.1The vector a

10、nd specify a natural spline g if and only if Then the roughness penalty will satisfy,Natural Cubic Splines,Value-second derivative representation,Natural Cubic Splines,Value-second derivative representation R is strictly diagonal dominant, i.e. R is positive definite, so we can define,Interpolating

11、Splines,To find a smooth curve that interpolate (ti,zi), i.e. g(ti)=zi for all i. Theorem 2.2Suppose and t1tn. Given any values z1,zn, there is a unique natural cubic spline g with knots ti satisfying,Interpolating Splines,The natural cubic spline interpolant is the unique minimizer of over S2a,b th

12、at interpolate the data. Theorem 2.3Suppose g is the interpolant natural cubic spline, then,Smoothing Splines,Penalized sum of squaresg: any twice-differentiable function on a,b : smoothing parameter (rate of exchange between residual error and local variation) Penalized least squares estimator,Smoo

13、thing Splines,1. The curve estimator is necessarily a natural cubic spline with knots at ti, for i=1,n. Proof: suppose g is the NCS,Smoothing Splines,2. Existence and uniqueness Let then since be precisely the vector of . Express ,Smoothing Splines,2. Theorem 2.4Let be the natural cubic spline with

14、knots at ti for which . Then for any in S2a,b,Smoothing Splines,3. The Reinsch algorithmThe matrix has bandwidth 5 and is symmetric and strictly positive-definite, therefore it has a Cholesky decomposition,Smoothing Splines,3. The Reinsch algorithm for spline smoothingStep 1: Evaluate the vector .St

15、ep 2: Find the non-zero diagonals of and hence the Cholesky decomposition factors L and D. Step 3: Solve for by forward and back substitution.Step 4: Find g by .,Smoothing Splines,4. Some concluding remarks Minimizing curve essentially does not depend on a and b, as long as all the data points lie b

16、etween a and b. If n=2, for any , setting to be the straight line through the two points (t1,Y1) and (t2,Y2) will reduce S(g) to zero. If n=1, the minimizer is no longer unique, since any straight line through (t1,Y1) will yield a zero value S(g).,Choosing the Smoothing Parameter,Two different philo

17、sophical approaches Subjective choice Automatic method chosen by data Cross-validation Generalized cross-validation,Choosing the Smoothing Parameter,Cross-validationGeneralized cross-validation,Available Software,smooth.spline in R Description:Fits a cubic smoothing spline to the supplied data. Usag

18、e: plot(speed, dist) cars.spl - smooth.spline(speed, dist) cars.spl2 - smooth.spline(speed, dist, df=10) lines(cars.spl, col = “blue“) lines(cars.spl2, lty=2, col = “red“),Available Software,Example 1library(modreg)y18 - c(1:3,5,4,7:3,2*(2:5),rep(10,4)xx - seq(1,length(y18), len=201)(s2 - smooth.spl

19、ine(y18) # GCV(s02 - smooth.spline(y18, spar = 0.2)plot(y18, main=deparse(s2$call), col.main=2) lines(s2, col = “blue“); lines(s02, col = “orange“); lines(predict(s2, xx), col = 2)lines(predict(s02, xx), col = 3); mtext(deparse(s02$call), col = 3),Available Software,Example 1,Available Software,Exam

20、ple 2data(cars) # N=50, n (# of distinct x) =19attach(cars)plot(speed, dist, main = “data(cars) & smoothing splines“)cars.spl df =“,round(cars.spl$df,1), “s( * , df = 10)“), col = c(“blue“,“red“), lty = 1:2, bg=bisque)detach(),Available Software,Example 2,Extensions of Roughness penalty approach,Sem

21、iparametric modeling: a simple application to multiple regressionGeneralized linear models (GLM) To allow all the explanatory variables to be nonlinearAdditive model approach,Reference,P.J. Green and B.W. Silverman (1994) Nonparametric Regression and Generalized Linear Models. London: Chapman & Hall,