- This event has passed.
Dr. Andreas Ziegler at the CIHR STAGE International Speaker Seminar Series
October 4, 2013 @ 12:00 pm - 2:00 pm
CIHR STAGE International Speaker Seminar Series proudly presents
Prof. Andreas Ziegler
Scientific Director, Center for Clinical Trials
University of Luebeck, Germany
Talk title: Probability Machines: Estimating individual probabilities using machine learning methods
Machine learning (ML) is increasingly used for data mining in genetics, biomedicine, credit scoring, weather forecasting, and other areas of application. Recent work has shown that machine learning can also be used for probability estimation by embedding the probability estimation problem in nonparametric regression estimation. As a result, nonparametric regression machines directly inherit their properties, such as consistency and convergence rate, to the corresponding probability machine. Their advantage over parametric standard statistical approaches, such as logistic regression is that probability machines do not require a correct specification of the functional relationship between the dependent variables and the independent variables. These methods provide robust nonparametric modeling of the regression function with minimal assumptions about the form of the relationships instead. Probability machines directly apply to assessing the probability of outcomes of interest based on different characteristics of individuals. In therapeutic observational studies, they can also be used for computing propensity scores for adjustment. They easily extend to dependent variables with multiple categories. In this contribution we first embed the probability estimation problem in nonparametric regression estimation. Next, we explore some consistent probability machines, such as random forest, k‐nearest neighbors, and bagged nearest neighbors for the purpose of probability estimation. We show how probabilities using probability machines can be estimated using standard software. Finally, we illustrate the approach using data, especially genetic data from the literature as well as from our own applications.
Dr. Ziegler earned his diploma in Statistics at Munich, Germany, in 1992, followed by a doctoral degree in Statistics at Dortmund, Germany, in 1994. In 1998, he completed his habilitation in Genetic Epidemiology and extended his venia legendi in 2000 by the subjects Medical Biometry and Epidemiology, both at the University of Marburg, Germany. He is professor and head of the Institute of Medical Biometry and Statistics at the University of Luebeck, Germany, since 2001. He is a board-certified biostatistician for controlled clinical trials since 1999. He was President of the German Region of the International Biometric Society, and member of the Presidium of the German Society of Medical Informatics, Biometry, and Epidemiology. He is President of the International Genetic Epidemiology Society and member of the Executive Board of the International Biometric Society. Ziegler has co-authored more than 400 articles and 9 books, including a textbook on Statistical Approaches to Genetic Epidemiology. Among other honours, he received the Susanne-Dahms medal for his contributions to the German Region of the International Biometric Society and several award for his scientific monographs and articles.