MLE_LambertW {LambertW}R Documentation

Maximum Likelihood Estimation

Description

Maximum Likelihood Estimation (MLE) theta for Lambert W RV for Gaussian and student-t input.

Usage

MLE_LambertW(y, distname = c("normal"), theta.0 = IGMM(y)$theta, hessian=TRUE)
## Default S3 method:
MLE_LambertW(y, distname = c("normal"), theta.0 = IGMM(y)$theta, hessian=TRUE)

Arguments

y a numeric vector of real values.
distname input distribution; default: "normal", alternative "t"
theta.0 starting value for numerical optimization; default: IGMM estimate.
hessian return the (numerically obtained) hessian matrix at the optimum?; default: TRUE

Value

An object of class LWest:

data the data
theta.0 initial value
theta MLE for theta
logLH log-likelihood function (argument for the summary function to numerically calculate the Hessian)
hessian Hessian matrix; used to calculate standard errors
call function call
message message from the optimization method. What kind of convergence?
distname character string indicating the input distribution: "t" or "normal".
method Estimation method. Here "MLE"

Author(s)

Georg M. Goerg

References

Goerg, G.M. (2009). “Lambert W Random Variables - A new class of skewed distribution functions”. Unpublished

Examples

data(AA)
attach(AA)

X=AA[AA$sex=="f",]
y=X$bmi

fit.ml=MLE_LambertW(y)
summary(fit.ml)
plot(fit.ml)

[Package LambertW version 0.1.9 Index]