Classical parametric estimation methods applied to nonlinear regression and limited-dependent-variable models are very sensitive to misspecification and data errors. On the other hand, semiparametric and nonparametric methods, which are not restricted by parametric assumptions, require more data...

We will study causal relationships of a known form between random variables. Given a model, we distinguish one or more dependent (endogenous) variables Y = (Y1, . . . , Yl), l ∈ N, which are explained by a model, and independent (exogenous, explanatory) variables X = (X1, . . . ,Xp), p ∈ N,...

Many methods of computational statistics lead to matrix-algebra or numerical- mathematics problems. For example, the least squares method in linear regression reduces to solving a system of linear equations. The principal components method is based on finding eigenvalues and eigenvectors of a...

Most dimension reduction methods based on nonparametric smoothing are highly sensitive to outliers and to data coming from heavy tailed distributions. We show that the recently proposed MAVE and OPG methods by Xia et al. (2002) allow us to make them robust in a relatively straightforward way...