new.lambda {pendensity}

Calculating new penalty parameter lambda

Description

Calculating new penalty parameter lambda.

Usage

new.lambda(penden.env,lambda0)

Arguments

penden.env

Containing all information, environment of pendensity()

lambda0

actual penalty parameter lambda

Details

Iterating for the lambda is stopped, when the changes between the old and the new lambda is smaller than 0.01*lambda0. If this criterion isn't reached, the iteration is terminated after 11 iterations.

The iteration formulae is

\lambda^{-1}=\frac{\hat{\beta}^T D_m \hat{\beta}}{df(\hat{\lambda})-(m-1)}.

Value

Returning the new lambda.

Author(s)

Christian Schellhase <[e-mail]>

References

Density Estimation with a Penalized Mixture Approach, Schellhase C. and Kauermann G. (2012), Computational Statistics 27 (4), p. 757-777.