Varying‐Thresholds Models: a Flexible Approach to Parametric and Nonparametric Regression
International Statistical Review · 2026-12 · 94 (3) · 608-634
DOI: 10.1111/insr.70018 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Statistical Review, online erschienen am 21.11.2025. Verfasst von Gerhard Tutz. Die Arbeit nennt 58 Literaturangaben.
Abstract (Originalsprache)
A general class of models is proposed that is able to estimate the whole predictive distribution of a dependent variable given a vector of explanatory variables . The models exploit that the strength of explanatory variables to distinguish between low and high values of the dependent variable may vary across the thresholds that are used to define low and high. Linear versions of the models are generalizations of classical linear regression models but also of widely used ordinal regression models. They allow to visualize the effect of explanatory variables in the form of parameter functions. More general models are based on efficient nonparametric approaches such as random forests, which are more flexible and are strong prediction tools. A general estimation method is given that can use all the estimation tools that have been proposed for binary regression, including selection methods like the lasso or elastic net. For linearly structured models, maximum likelihood estimates are derived. The usefulness of the models is illustrated by simulations and several real data set.
Zitieren
APA
Tutz, G. (2026). Varying‐Thresholds Models: a Flexible Approach to Parametric and Nonparametric Regression. International Statistical Review, 94(3), 608-634. https://doi.org/10.1111/insr.70018
Vancouver
Tutz G. Varying‐Thresholds Models: a Flexible Approach to Parametric and Nonparametric Regression. International Statistical Review. 2026;94(3):608-634. doi:10.1111/insr.70018
BibTeX
@article{Tutz2026,
author = {Tutz, Gerhard},
title = {Varying‐Thresholds Models: a Flexible Approach to Parametric and Nonparametric Regression},
journal = {International Statistical Review},
year = {2026},
volume = {94},
number = {3},
pages = {608--634},
doi = {10.1111/insr.70018}
}
EUVIMED