hide long namesshow long names
hide short namesshow short names
Integer type:  int32  int64  nag_int  show int32  show int32  show int64  show int64  show nag_int  show nag_int

PDF version (NAG web site, 64-bit version, 64-bit version)
Chapter Contents
Chapter Introduction
NAG Toolbox

NAG Toolbox: nag_univar_robust_1var_mestim (g07db)


    1  Purpose
    2  Syntax
    7  Accuracy
    9  Example


nag_univar_robust_1var_mestim (g07db) computes an M-estimate of location with (optional) simultaneous estimation of the scale using Huber's algorithm.


[theta, sigma, rs, nit, wrk, ifail] = g07db(isigma, x, ipsi, c, h1, h2, h3, dchi, theta, sigma, tol, 'n', n, 'maxit', maxit)
[theta, sigma, rs, nit, wrk, ifail] = nag_univar_robust_1var_mestim(isigma, x, ipsi, c, h1, h2, h3, dchi, theta, sigma, tol, 'n', n, 'maxit', maxit)


The data consists of a sample of size n, denoted by x1,x2,,xn, drawn from a random variable X.
The xi are assumed to be independent with an unknown distribution function of the form
where θ is a location argument, and σ is a scale argument. M-estimators of θ and σ are given by the solution to the following system of equations:
i=1nψ xi-θ^ /σ^=0 (1)
i=1nχ xi-θ^ /σ^=n-1β (2)
where ψ and χ are given functions, and β is a constant, such that σ^ is an unbiased estimator when xi, for i=1,2,,n has a Normal distribution. Optionally, the second equation can be omitted and the first equation is solved for θ^ using an assigned value of σ=σc.
The values of ψ xi-θ^σ^ σ^ are known as the Winsorized residuals.
The following functions are available for ψ and χ in nag_univar_robust_1var_mestim (g07db).
(a) Null Weights
ψt=t t22   χt= t22
Use of these null functions leads to the mean and standard deviation of the data.
(b) Huber's Function
ψt = max-c,minc,t t22   χt= t22 td
    χt= d22 t>d
(c) Hampel's Piecewise Linear Function
ψh1,h2,h3t=t t22 0th1 t22 χt= t22 td
ψh1,h2,h3t=h1 h1th2  
ψh1,h2,h3t=h1h3-t/h3-h2 t22 h2th3 t22 χt= d22 t>d
ψh1,h2,h3t=0 t>h3  
(d) Andrew's Sine Wave Function
ψt=sint d22 -πtπ d22 χt= t22 td
ψt=0 d22 otherwise d22 χt= d22 t>d
(e) Tukey's Bi-weight
ψt=t 1-t2 2 t22 t1 t22 χt= t22 td
ψt=t 1-t2 2=0 t22 otherwise t22 χt= d22 t>d
where c, h1, h2, h3 and d are constants.
Equations (1) and (2) are solved by a simple iterative procedure suggested by Huber:
σ^k=1βn-1 i=1nχ xi-θ^k-1σ^k-1 σ^k-12  
θ^k=θ^k- 1+1ni= 1nψ xi-θ^k- 1σ^k σ^k  
σ^k=σc,   if  σ  is fixed.  
The initial values for θ^ and σ^ may either be user-supplied or calculated within nag_univar_robust_1var_mestim (g07db) as the sample median and an estimate of σ based on the median absolute deviation respectively.
nag_univar_robust_1var_mestim (g07db) is based upon function LYHALG within the ROBETH library, see Marazzi (1987).


Hampel F R, Ronchetti E M, Rousseeuw P J and Stahel W A (1986) Robust Statistics. The Approach Based on Influence Functions Wiley
Huber P J (1981) Robust Statistics Wiley
Marazzi A (1987) Subroutines for robust estimation of location and scale in ROBETH Cah. Rech. Doc. IUMSP, No. 3 ROB 1 Institut Universitaire de Médecine Sociale et Préventive, Lausanne


Compulsory Input Parameters

1:     isigma int64int32nag_int scalar
The value assigned to isigma determines whether σ^ is to be simultaneously estimated.
The estimation of σ^ is bypassed and sigma is set equal to σc.
σ^ is estimated simultaneously.
2:     xn – double array
The vector of observations, x1,x2,,xn.
3:     ipsi int64int32nag_int scalar
Which ψ function is to be used.
Huber's function.
Hampel's piecewise linear function.
Andrew's sine wave,
Tukey's bi-weight.
4:     c – double scalar
If ipsi=1, c must specify the argument, c, of Huber's ψ function. c is not referenced if ipsi1.
Constraint: if ipsi=1, c>0.0.
5:     h1 – double scalar
6:     h2 – double scalar
7:     h3 – double scalar
If ipsi=2, h1, h2 and h3 must specify the arguments, h1, h2, and h3, of Hampel's piecewise linear ψ function. h1, h2 and h3 are not referenced if ipsi2.
Constraint: 0h1h2h3 and h3>0.0 if ipsi=2.
8:     dchi – double scalar
d, the argument of the χ function. dchi is not referenced if ipsi=0.
Constraint: if ipsi0, dchi>0.0.
9:     theta – double scalar
If sigma>0 then theta must be set to the required starting value of the estimation of the location argument θ^. A reasonable initial value for θ^ will often be the sample mean or median.
10:   sigma – double scalar
The role of sigma depends on the value assigned to isigma, as follows:
  • if isigma=1, sigma must be assigned a value which determines the values of the starting points for the calculations of θ^ and σ^. If sigma0.0 then nag_univar_robust_1var_mestim (g07db) will determine the starting points of θ^ and σ^. Otherwise the value assigned to sigma will be taken as the starting point for σ^, and theta must be assigned a value before entry, see above;
  • if isigma=0, sigma must be assigned a value which determines the value of σc, which is held fixed during the iterations, and the starting value for the calculation of θ^. If sigma0, then nag_univar_robust_1var_mestim (g07db) will determine the value of σc as the median absolute deviation adjusted to reduce bias (see nag_univar_robust_1var_median (g07da)) and the starting point for θ^. Otherwise, the value assigned to sigma will be taken as the value of σc and theta must be assigned a relevant value before entry, see above.
11:   tol – double scalar
The relative precision for the final estimates. Convergence is assumed when the increments for theta, and sigma are less than tol×max1.0,σk-1.
Constraint: tol>0.0.

Optional Input Parameters

1:     n int64int32nag_int scalar
Default: the dimension of the array x.
n, the number of observations.
Constraint: n>1.
2:     maxit int64int32nag_int scalar
Default: 50
The maximum number of iterations that should be used during the estimation.
Constraint: maxit>0.

Output Parameters

1:     theta – double scalar
The M-estimate of the location argument, θ^.
2:     sigma – double scalar
Contains the M-estimate of the scale argument, σ^, if isigma was assigned the value 1 on entry, otherwise sigma will contain the initial fixed value σc.
3:     rsn – double array
The Winsorized residuals.
4:     nit int64int32nag_int scalar
The number of iterations that were used during the estimation.
5:     wrkn – double array
If sigma0.0 on entry, wrk will contain the n observations in ascending order.
6:     ifail int64int32nag_int scalar
ifail=0 unless the function detects an error (see Error Indicators and Warnings).

Error Indicators and Warnings

Errors or warnings detected by the function:
On entry,n1,
orisigma0 or 1,
On entry,c0.0 and ipsi=1,
orh1<0.0 and ipsi=2,
orh1=h2=h3=0.0 and ipsi=2,
orh1>h2 and ipsi=2,
orh1>h3 and ipsi=2,
orh2>h3 and ipsi=2,
ordchi0.0 and ipsi0.
On entry,all elements of the input array x are equal.
sigma, the current estimate of σ, is zero or negative. This error exit is very unlikely, although it may be caused by too large an initial value of sigma.
The number of iterations required exceeds maxit.
On completion of the iterations, the Winsorized residuals were all zero. This may occur when using the isigma=0 option with a redescending ψ function, i.e., Hampel's piecewise linear function, Andrew's sine wave, and Tukey's biweight.
If the given value of σ is too small, then the standardized residuals xi-θ^kσc , will be large and all the residuals may fall into the region for which ψt=0. This may incorrectly terminate the iterations thus making theta and sigma invalid.
Re-enter the function with a larger value of σc or with isigma=1.
An unexpected error has been triggered by this routine. Please contact NAG.
Your licence key may have expired or may not have been installed correctly.
Dynamic memory allocation failed.


On successful exit the accuracy of the results is related to the value of tol, see Arguments.

Further Comments

When you supply the initial values, care has to be taken over the choice of the initial value of σ. If too small a value of σ is chosen then initial values of the standardized residuals xi-θ^kσ  will be large. If the redescending ψ functions are used, i.e., Hampel's piecewise linear function, Andrew's sine wave, or Tukey's bi-weight, then these large values of the standardized residuals are Winsorized as zero. If a sufficient number of the residuals fall into this category then a false solution may be returned, see page 152 of Hampel et al. (1986).


The following program reads in a set of data consisting of eleven observations of a variable X.
For this example, Hampel's Piecewise Linear Function is used (ipsi=2), values for h1, h2 and h3 along with d for the χ function, being read from the data file.
Using the following starting values various estimates of θ and σ are calculated and printed along with the number of iterations used:
(a) nag_univar_robust_1var_mestim (g07db) determines the starting values, σ is estimated simultaneously.
(b) You must supply the starting values, σ is estimated simultaneously.
(c) nag_univar_robust_1var_mestim (g07db) determines the starting values, σ is fixed.
(d) You must supply the starting values, σ is fixed.
function g07db_example

fprintf('g07db example results\n\n');

x = [13; 11; 16;  5;  3; 18;  9;  8;  6; 27;  7];

ipsi = int64(2);
c  = 0;
h1 = 1.5;
h2 = 3;
h3 = 4.5;
dchi = 1.5;
tol = 0.0001;

% Loop over input values for isigma sigma and theta
isigma = int64([ 1  1  0  0]);
sigma  =         [-1  7 -1  7];
theta  =         [ 0  2  0  2];

fprintf('           Input parameters     Output parameters\n');
fprintf(' isigma   sigma   theta   tol    sigma  theta\n');
for j = 1:numel(theta)

  fprintf('%3d   %8.4f%8.4f%8.4f', isigma(j), sigma(j), theta(j), tol);

  [thetaOut, sigmaOut, rs, nit, wrk, ifail] = ...
  g07db( ...
         isigma(j), x, ipsi, c, h1, h2, h3, dchi, theta(j), sigma(j), tol);

  fprintf(' %8.4f%8.4f\n', sigmaOut, thetaOut);

g07db example results

           Input parameters     Output parameters
 isigma   sigma   theta   tol    sigma  theta
  1    -1.0000  0.0000  0.0001   6.3247 10.5487
  1     7.0000  2.0000  0.0001   6.3249 10.5487
  0    -1.0000  0.0000  0.0001   5.9304 10.4896
  0     7.0000  2.0000  0.0001   7.0000 10.6500

PDF version (NAG web site, 64-bit version, 64-bit version)
Chapter Contents
Chapter Introduction
NAG Toolbox

© The Numerical Algorithms Group Ltd, Oxford, UK. 2009–2015