statmod (version 1.5.0)

mscale: M Scale Estimation

Description

Robust estimation of a scale parameter using Hampel's redescending psi function.

Usage

mscale(u, na.rm=FALSE)

Value

numeric constant giving the estimated scale.

Arguments

u

numeric vector of residuals.

na.rm

logical. Should missing values be removed?

Author

Gordon Smyth

Details

Estimates a scale parameter or standard deviation using an M-estimator with 50% breakdown. This means the estimator is highly robust to outliers. If the input residuals u are a normal sample, then mscale(u) should be equal to the standard deviation.

References

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.

Examples

u <- rnorm(100)
sd(u)
mscale(u)