We propose a new information theoretically based optimization criterion for the estimation of mixture density models and compare it with other methods based on maximum likelihood and maximum a posterio estimation. For the optimization, we employ an evolutionary algorithm which estimates both...

Sliced Inverse Regression (SIR) is a promising technique for the purpose of dimension reduction. Several properties of this relatively new method have been examined already, but little attention has been paid to robustness aspects. We show that SIR is very sensitive towards outliers in the data....

In this note several aspects of specification tests in nonparametric models driven by an absolutely regular process are discussed, which were recently proprosed in the literature. In particular we give a more detailed asymptotic analysis of tests based on kernel methods under fixed alternatives...

This paper shows that the properties of nonlinear transformations of a fractionally integrated process depend strongly on whether the initial series is stationary or not. Transforming a stationary Gaussian I(d) process with d 0 leads to a long-memory process with the same or a smaller...

The aim of this paper is to find a modeling approach for spatially and temporally structured data. The spatial distribution is considered to form an irregular lattice with a specified definition of neighborhood. Additional to the spatial component, a temporal autoregressive parameter, and a time...

Conditions for the consistency of the estimator s2 of the variance of the disturbance a2u under first-order spatial error processes are given.

In the presence of variability control factors in Taguchi experiments, then the original b-method (Logothetis 1990) is liable to lead to wrong transformations. We propose a generalization of the b-method which should lead to correct transformations, even if there is a variability control factor...