nag_smooth_spline_estim (g10acc) (PDF version)
g10 Chapter Contents
g10 Chapter Introduction
NAG Library Manual

NAG Library Function Document

nag_smooth_spline_estim (g10acc)


    1  Purpose
    7  Accuracy

1  Purpose

nag_smooth_spline_estim (g10acc) estimates the values of the smoothing argument and fits a cubic smoothing spline to a set of data.

2  Specification

#include <nag.h>
#include <nagg10.h>
void  nag_smooth_spline_estim (Nag_SmoothParamMethods method, Integer n, const double x[], const double y[], const double weights[], double yhat[], double coeff[], double *rss, double *df, double res[], double h[], double *crit, double *rho, double u, double tol, Integer maxcal, NagError *fail)

3  Description

For a set of n  observations x i , y i , for i=1,2,,n, the spline provides a flexible smooth function for situations in which a simple polynomial or nonlinear regression model is not suitable.
Cubic smoothing splines arise as the unique real-valued solution function, f , with absolutely continuous first derivative and squared-integrable second derivative, which minimizes:
i=1 n w i y i - f x i 2 + ρ - f x 2 dx ,  
where w i  is the (optional) weight for the i th observation and ρ  is the smoothing argument. This criterion consists of two parts: the first measures the fit of the curve and the second the smoothness of the curve. The value of the smoothing argument ρ  weights these two aspects; larger values of ρ  give a smoother fitted curve but, in general, a poorer fit. For details of how the cubic spline can be fitted see Hutchinson and de Hoog (1985) and Reinsch (1967).
The fitted values, y ^ = y ^ 1 , y ^ 2 , , y ^ n T , and weighted residuals, r i , can be written as:
y ^ = Hy  and  r i = w i y i - y ^ i  
for a matrix H . The residual degrees of freedom for the spline is trace I-H  and the diagonal elements of H  are the leverages.
The argument ρ  can be estimated in a number of ways.
1. The degrees of freedom for the spline can be specified, i.e., find ρ  such that trace H = ν 0  for given ν 0 .
2. Minimize the cross-validation (CV), i.e., find ρ  such that the CV is minimized, where
CV = 1 i=1 n w i i=1 n r i 1 - h ii 2 .  
3. Minimize the generalized cross-validation (GCV), i.e., find ρ  such that the GCV is minimized, where
GCV = n 2 i=1 n w i i=1 n r i 2 i=1 n 1 - h ii 2 .  
nag_smooth_spline_estim (g10acc) requires the x i  to be strictly increasing. If two or more observations have the same x i  value then they should be replaced by a single observation with y i  equal to the (weighted) mean of the y  values and weight, w i , equal to the sum of the weights. This operation can be performed by nag_order_data (g10zac).
The algorithm is based on Hutchinson (1986).

4  References

Hastie T J and Tibshirani R J (1990) Generalized Additive Models Chapman and Hall
Hutchinson M F (1986) Algorithm 642: A fast procedure for calculating minimum cross-validation cubic smoothing splines ACM Trans. Math. Software 12 150–153
Hutchinson M F and de Hoog F R (1985) Smoothing noisy data with spline functions Numer. Math. 47 99–106
Reinsch C H (1967) Smoothing by spline functions Numer. Math. 10 177–183

5  Arguments

1:     method Nag_SmoothParamMethodsInput
On entry: indicates whether the smoothing argument is to be found by minimization of the CV or GCV functions, or by finding the smoothing argument corresponding to a specified degrees of freedom value.
Cross-validation is used.
The degrees of freedom are specified.
Generalized cross-validation is used.
Constraint: method=Nag_SmoothParamCV, Nag_SmoothParamDF or Nag_SmoothParamGCV.
2:     n IntegerInput
On entry: the number of observations, n .
Constraint: n3 .
3:     x[n] const doubleInput
On entry: the distinct and ordered values x i , for i=1,2,,n.
Constraint: x[i-1] < x[i] , for i=1,2,,n-1.
4:     y[n] const doubleInput
On entry: the values y i , for i=1,2,,n.
5:     weights[n] const doubleInput
On entry: weights must contain the n  weights, if they are required. Otherwise, weights must be set to NULL.
Constraint: if weights are required, then weights[i-1] > 0.0 , for i=1,2,,n.
6:     yhat[n] doubleOutput
On exit: the fitted values, y ^ i , for i=1,2,,n.
7:     coeff[n-1×3] doubleOutput
On exit: the spline coefficients. More precisely, the value of the spline approximation at t  is given by coeff[ i-1 × n-1 + 2 ] × d + coeff[ i-1 × n-1 + 1 ] × d + coeff[ i-1 × n-1 ] × d + y ^ i , where x i t < x i+1  and d = t-x i .
8:     rss double *Output
On exit: the (weighted) residual sum of squares.
9:     df double *Output
On exit: the residual degrees of freedom. If method=Nag_SmoothParamDF, this will be n-crit  to the required accuracy.
10:   res[n] doubleOutput
On exit: the (weighted) residuals, r i , for i=1,2,,n.
11:   h[n] doubleOutput
On exit: the leverages, h ii , for i=1,2,,n.
12:   crit double *Input/Output
On entry: if method=Nag_SmoothParamDF, the required degrees of freedom for the spline.
If method=Nag_SmoothParamCV or Nag_SmoothParamGCV, crit need not be set.
Constraint: 2.0 < crit n .
On exit: if method=Nag_SmoothParamCV, the value of the cross-validation, or if method=Nag_SmoothParamGCV, the value of the generalized cross-validation function, evaluated at the value of ρ  returned in rho.
13:   rho double *Output
On exit: the smoothing argument, ρ .
14:   u doubleInput
On entry: the upper bound on the smoothing argument. See Section 9 for details on how this argument is used.
Suggested value: u=1000.0 .
Constraint: u>tol .
15:   tol doubleInput
On entry: the accuracy to which the smoothing argument rho is required. tol should be preferably not much less than ε , where ε  is the machine precision.
Constraint: tol machine precision .
16:   maxcal IntegerInput
On entry: the maximum number of spline evaluations to be used in finding the value of ρ .
Suggested value: maxcal=30 .
Constraint: maxcal3 .
17:   fail NagError *Input/Output
The NAG error argument (see Section 3.6 in the Essential Introduction).

6  Error Indicators and Warnings

On entry, u=value  while tol=value . These arguments must satisfy u>tol .
Dynamic memory allocation failed.
On entry, argument method had an illegal value.
A solution to the accuracy given by tol has not been achieved in maxcal iterations. Try increasing the value of tol and/or maxcal.
method=Nag_SmoothParamCV or Nag_SmoothParamGCV and the optimal value of rho>u . Try a larger value of u.
method=Nag_SmoothParamDF and the required value of rho for specified degrees of freedom > u . Try a larger value of u.
method=Nag_SmoothParamDF and the accuracy given by tol cannot be achieved. Try increasing the value of tol.
On entry, maxcal=value.
Constraint: maxcal3.
On entry, n=value.
Constraint: n3.
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
The sequence x is not strictly increasing: x[value] = value, x[value] = value.
On entry, crit=value .
Constraint: crit>2.0 , if method=Nag_SmoothParamDF.
On entry, weights[value] = value.
Constraint: weights[i] > 0 , for i=0,1,,n - 1.
On entry, crit=value  and n=value . These arguments must satisfy critn , if method=Nag_SmoothParamDF.
On entry, tol=value , machine precision nag_machine_precision = value.
Constraint: tolmachine precision .

7  Accuracy

When minimizing the cross-validation or generalized cross-validation, the error in the estimate of ρ  should be within ± 3 × tol × rho + tol . When finding ρ  for a fixed number of degrees of freedom the error in the estimate of ρ  should be within ± 2 × tol × max1,rho .
Given the value of ρ , the accuracy of the fitted spline depends on the value of ρ  and the position of the x  values. The values of x i - x i-1  and w i  are scaled and ρ  is transformed to avoid underflow and overflow problems.

8  Parallelism and Performance

Not applicable.

9  Further Comments

The time to fit the spline for a given value of ρ  is of order n .
When finding the value of ρ  that gives the required degrees of freedom, the algorithm examines the interval 0.0 to u. For small degrees of freedom the value of ρ  can be large, as in the theoretical case of two degrees of freedom when the spline reduces to a straight line and ρ  is infinite. If the CV or GCV is to be minimized then the algorithm searches for the minimum value in the interval 0.0 to u. If the function is decreasing in that range then the boundary value of u will be returned. In either case, the larger the value of u the more likely is the interval to contain the required solution, but the process will be less efficient.
Regression splines with a small <n  number of knots can be fitted by nag_1d_spline_fit_knots (e02bac) and nag_1d_spline_fit (e02bec).

10  Example

The data, given by Hastie and Tibshirani (1990), is the age, x i , and C-peptide concentration (pmol/ml), y i , from a study of the factors affecting insulin-dependent diabetes mellitus in children. The data is input, reduced to a strictly ordered set by nag_order_data (g10zac) and a spline with 5 degrees of freedom is fitted by nag_smooth_spline_estim (g10acc). The fitted values and residuals are printed.

10.1  Program Text

Program Text (g10acce.c)

10.2  Program Data

Program Data (g10acce.d)

10.3  Program Results

Program Results (g10acce.r)

nag_smooth_spline_estim (g10acc) (PDF version)
g10 Chapter Contents
g10 Chapter Introduction
NAG Library Manual

© The Numerical Algorithms Group Ltd, Oxford, UK. 2015