nlr (version 0.1-3)

parInfer.WM: WM-estimate Inference

Description

Parameter inference for weighted M-estimate. WM-estimate is based on minimizing the robustified form of likelihood, simultanously over nonlinear function parameter and variance model parameters the the covariance of parameter, the estimate is assymptotically normal (See Lim et al.2010) with given covariance matric which compute for sample by the function.

Usage

parInfer.WM(object, confidence = 0.95)

Arguments

object

nl.fitt.rgn object of WM-fitt generated by nl.robhetroWM function.

confidence

Confidence probability.

Value

covmat:

Covariance matrix of nonlinear model function parameters.

covtau

Covariance matrix of nonlinear variance model parameters.

parstdev

Standard deviation of nonlinear model function parameter. It is square root of diagonal of covmat.

CI

Confidence interval for nonlinear model function parameter.

Details

Compute covariance matrix and confidence interval for nonlinear model function parameter and nonlinear variance model parameters.

References

Lim, C., Sen, P. K., Peddada, S. D. (2010). Statistical inference in nonlinear regression under heteroscedasticity. Sankhya B 72:202-218.

See Also

nl.fitt.rgn, nl.robhetroWM

Examples

# NOT RUN {
## The function is currently defined as
"parInfer.WM"
# }