sqrtvat: Compute square root of vairiance attribute.

In nlr: Nonlinear Regression Modelling using Robust Methods

Description

Transform variance to standard deviation with all its gradient and hessian.

Usage

sqrtvat(varcomp)

Arguments

varcomp

Is (n*1) vector of some variance, transform to √ (vc), with attributes attr(vc,"gradient"), n \times p gradient. And attr(vc,"hessian"), n \times p \ times{p} hessian.

Details

For computation purpose to transform variance function values to standard deviation function value is used.

Value

Standard deviation is equal the square root of variance, with Gradient equal to:

Gradient (sdev) = \frac{1}{2} √{Var} \times Gradient (Var)

and hessian is equal

hessian(sdev) = \frac{1}{2} √{vc} \times hesian(vc) - (\frac{1}{4} σ ^ 3) grad(vc)^T \%m3d\% grad(vc)

Note

Is used for when standard deviation of a heterogeneous variance function model is needed.

This function call by nlr, for compatibility it is better to call from nlr rather than directly by user.

Author(s)

Hossein Riazoshams, May 2014. Email: [e-mail] URL http://www.riazoshams.com/nlr/

References

Riazoshams H, Midi H, and Ghilagaber G, 2018,. Robust Nonlinear Regression, with Application using R, Joh Wiley and Sons.

See Also

nlr

Examples

## The function is currently defined as
"sqrtvat"

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