RCig {ig}R Documentation

Relative change on the MLE of the IGTD

Description

The function RCig computes the relative change (RC) on the MLE when some observations are removed in order to evaluate the effect of their potential influence.

Usage

RCig(x, cases.removed = NULL, kernel = "normal")

Arguments

x Vector of observations.
cases.removed Index of the potentially influential case(s) that must be removed.
kernel Kernel of the pdf of the associated symmetrical distribution by means of which the IGTD is obtained. The kernels: "Laplace", "logistic", "normal" and "t" are available.

Details

This function computes the relative changes (RC), in percentage, of each estimated parameter, defined by mbox{RC}_{theta_j} = |(hat theta_j -hat theta_{j(mbox{tiny I})})/hat theta_j| times 100%, where hat theta_{j(mbox{tiny I})} denotes the MLE of theta_j after the set I of cases has been removed.

Value

RCig() gives the RC on the MLE of the parameters of the IGTD from a sample of observations without to consider some potentially influential case(s) related to the MLE of the complete sample.

Author(s)

Víctor Leiva <victor.leiva@uv.cl>, Hugo Hernández <hugo.hernandez@msn.com>, and Antonio Sanhueza <asanhue@ufro.cl>.

References

Sanhueza, A., Leiva, V. and Balakrishnan, N. (2007). A new class of inverse Gaussian type distributions. Metrika (in press).

Examples

## Generates a sample from the IGTD
## Computes the RC the MLE of the parameters of the IGTD with g="normal" when the
##  case 1 has been removed
x<-rig(300,mu=1,lambda=1,kernel="normal")
RCig(x,cases.removed=c(1),kernel="normal")

[Package ig version 1.0 Index]