jaclev: Jacobian Leverage for nonlinear regression.

In nlr: Nonlinear Regression Modelling using Robust Methods

Description

Compute the Jacobian Leverage, generalized for nonlinear case.

Usage

jaclev(gradient, hessian, rsd)

Arguments

gradient	n \times p gradient of nonlinear function.
hessian	three simentional n \times p \times p of hessian of nonlinear regression function.
rsd	n \times 1 residual vector.

Details

Jacobian leverage, generalized form of hat matrix for nonlinear regression.

Value

n \times n matrix of jacobian leverages.

Note

Jacobian leverage for nonlinear regression is direct definition of perturbing response, thus it is free from the problems due to linear approximation of nonlinear function.

Author(s)

Laurent. R. T. ST., and Cook.

References

Laurent. R. T. ST., and Cook. R. D. (1992). Leverage and Superleverage in Nonlinear Regression, Journal of the American Statistical Association 87(420): 985-990.

See Also

nl.fitt, nl.fitt.gn

Examples

## The function is currently defined as
"jaclev"

nlr documentation built on July 31, 2019, 5:09 p.m.