In multinomial logit models, the identifiability of parameter estimates is typically obtained by side constraints that specify one of the response categories as reference category. When parameters are penalized, shrinkage of estimates should not depend on the reference category. In this paper we...

Rating scales as predictors in regression models are typically treated as metrically scaled variables or, alternatively, are coded in dummy variables. The first approach implies a scale level that is not justified, the latter approach results in a large number of parameters to be estimated....

The use of the multinomial logit model is typically restricted to applications with few predictors, because in high-dimensional settings maximum likelihood estimates tend to deteriorate. A sparsity-inducing penalty is proposed that accounts for the special structure of multinomial models by...

In this paper R2-type measures of the explanatory power of multivariate linear and categorical probit models proposed in the literature are reviewed and their deficiencies are discussed. It is argued that a measure of the explanatory power should take into account the components which are...

Additive models of the type y=f_1(x_1)+...+f_p(x_p)+e where f_j,j=1,...,p, have unspecified functional form, are flexible statistical regression models which can be used to characterize nonlinear regression effects. The basic tools used for fitting the additive model are the expansion in...

Various supervised learning and gene selection methods have been used for cancer diagnosis. Most of these methods do not consider interactions between genes, although this might be interesting biologically and improve classification accuracy. Here we introduce a new CART-based method to discover...

A shrinkage type estimator is introduced which has favorable properties in binary regression. Although binary observations are never very far away from the underlying probability, in all interesting cases there is a non-zero distance between observation and underlying mean. The proposed response...

Principal components are a well established tool in dimension reduction. The extension to principal curves allows for general smooth curves which pass through the middle of a p-dimensional data cloud. In this paper local principal curves are introduced, which are based on the localization of...