/ Home
S-Archive Download Script
mmnl MM Estimation for Nonlinear Regression
Estimates a nonlinear regression equation using 95% efficient MM estimation. User supplies the regression function and its gradient matrix.
X numeric design matrix. Number of rows must agree with y.
y numeric vector of observations.
b numeric vector of starting values for the parameters.
fun function giving vector of fitted values as a function of X and b.
grad function giving gradient matrix as a function of X and b.
scale numeric constant giving scale of the residuals. By default this is mscale(y - fun(X,b)).
trace logical constant. If true, parameters and criterion are printed at each iteration.
b numeric vector of estimated parametes.
fitted numeric vector of same length as y of fitted values.
residuals numeric vector of same length as y of residuals.
scale numeric constant giving the estimated or input scale.
criterion numeric constant giving the estimated criterion functoin.
Uses Hampel's redescending psi function. The estimators simultaneously have high breakdown and 95% efficiency under normal errors given consistent high breakdown starting values.
Yohai, V. J. (1987). High breakdown point and high efficiency robust estimates for regression. Ann. Statist. 15, 642-656.

Stromberg, A. J. (1993). Computation of high breakdown nonlinear regression parameters. J. Amer. Statist. Assoc. 88, 237-244.

Smyth, G. K., and Hawkins, D. M. (2000). Robust frequency estimation using elemental sets. Journal of Computational and Graphical Statistics 9, 196-214. (Abstract - Zipped PostScript)
mscale, rho.hampel, psi.hampel, mmfreq, robfreq
The function mmfreq gives an example of using mmnl.


S-Archive Download Script

Gordon Smyth. Copyright © 1996-2016. Last modified: 10 February 2004