Archive / INF Seminars / INF_2022_11_29_Ziegel_Johanna
USI - Email
 
 
Università
della
Svizzera
italiana
INF
 
 
 
  
 main_banner
 

Distributional (Single) Index Models

 
 
 

Host: Prof. Paul Schneider

 

Tuesday

29.11

USI Campus Est, room D1.13, Sector D // Online on Zoom
12:30 - 14:00
  
 

Johanna Ziegel
University of Bern
Click here to Join

Abstract: A Distributional (Single) Index Model (DIM) is a semiparametric model for distributional regression, that is, estimation of conditional distributions given covariates. The method is a combination of classical single- index models for the estimation of the conditional mean of a response given covariates, and isotonic distributional regression. The model for the index is parametric, whereas the conditional distributions are estimated nonparametrically under a stochastic ordering constraint. We show consistency of our estimators and apply them to a highly challenging dataset on the length of stay (LoS) of patients in intensive care units. We use the model to provide skillful and calibrated probabilistic predictions for the LoS of individual patients, which outperform the available methods in the literature.

Biography: Johanna Ziegel is Associate Professor of Applied Stochastics at the University of Bern (UniBe), Switzerland, and Visiting Scientist at the Heidelberg Institute for Theoretical Studies, Germany. Currently, she is the Director of the Institute of Mathematical Statistics and Actuarial Science, Head of the continuing education program Statistical Data Science, and Scientific Advisory Board Member of the Oeschger Centre for Climate Change Research at UniBe. Johanna obtained her PhD in 2009 at ETH Zurich, Switzerland. She held postdoctoral positions at the University of Melbourne, Australia, and at Heidelberg University, Germany before joining UniBe in 2012 as an Assistant Professor tenure track. She is an elected ISI member and has been a Council Member of the Bernoulli Society. Currently she is part of the editorial board of Bernoulli, JASA: Theory & Methods, and the International Journal of Forecasting. In 2021/2022, she has been part of the H.I.T. Program: High Potential University Leaders Identity & Skills Training Program – Inclusive Leadership in Academia. In 2022, she has won the Credit Swiss Award for Best Teaching at UniBe.