psiandrew function

Andrew's Sine Function

Andrew's sine function for use when fitting a linear model by robust regression using an M-estimator.

psi.andrew(u, k=1.339, deriv=0)

Arguments

• u: Numeric vector of evaluation points.
• k: Tuning constant. The suggested default value is 1.339.
• deriv: 0 or 1: to compute values of this function or of its first derivative. ## Returns

psi.andrew returns a vector of points evaluated using Andrew's sine function. ## See Also

rlm

References

Andrew, D. F. (1974), A Robust Method for Multiple Linear Regression, Technometrics, 16 , 523--531.

Young, D. S. (2017), Handbook of Regression Methods, CRC Press.

Examples

## Robust fit of the stackloss dataset.

require(MASS)

data(stackloss, package="datasets")

out <- rlm(stack.loss ~ ., data = stackloss, 
           psi = psi.andrew)

out