Vorlesen

Research Seminar Series in Statistics and Mathematics

Wirtschaftsuniversität Wien, Departments 4 D4.4.00810:30 - 11:45

Art Vortrag/Diskussion
SpracheEnglish
Vortragende/rTorsten Hothorn (Epidemiology, Biostatistics and Prevention Institute, University of Zurich)
Veranstalter Institut für Statistik und Mathematik
Kontakt katrin.artner@wu.ac.at

Torsten Hothorn (Epidemiology, Biostatistics and Prevention Institute, University of Zurich) about "Transformation Forests"

The Institute for Statistics and Mathematics (Department of Finance, Accounting and Statistics) cordially invites everyone interested to attend the talks in our Research Seminar Series, where internationally renowned scholars from leading universities present and discuss their (working) papers.

The list of talks for the winter term 2018/19 is available via the following link: https://www.wu.ac.at/en/statmath/resseminar

Abstract:
Regression models for supervised learning problems with a continuous response are commonly understood as models for the conditional mean of the response given predictors. This notion is simple and therefore appealing for interpretation and visualisation. Information about the whole underlying conditional distribution is, however, not available from these models. A more general understanding of regression models as models for conditional distributions allows much broader inference from such models, for example the computation of prediction intervals. Several random forest-type algorithms aim at estimating conditional distributions, most prominently quantile regression forests (Meinshausen, 2006, JMLR). We propose a novel approach based on a parametric family of distributions characterised by their transformation function. A dedicated novel “transformation tree” algorithm able to detect distributional changes is developed. Based on these transformation trees, we introduce “transformation forests” as an adaptive local likelihood estimator of conditional distribution functions. The resulting predictive distributions are fully parametric yet very general and allow inference procedures, such as likelihood-based variable importances, to be applied in a straightforward way. The procedure allows general transformation models to be estimated without the necessity of a priori specifying the dependency structure of parameters. Applications include the computation of probabilistic forecasts, modelling differential treatment effects, or the derivation of counterfactural distributions for all types of response variables.

Kindly note that on October 12 two talks are scheduled at our institute:
9:00  Walter Farkas
10:30  Torsten Hothorn



zurück zur Übersicht