Christian Hennig: some recent presentations
Philosophy and foundations of statistics
I have presented much more than I have managed to publish yet of my ideas
on the philosophy and foundations of statistics. So here is some material.
- The role of model assumptions in statistics. Darmstadt, June 2017.
- Video of the RSS discussion meeting on Beyond subjective and objective in statistics (joint work with Andrew Gelman), London, April 2017.
- Human decisions in machine learning and statistical data analysis - always required, often ignored. Brussels, April 2017.
- Beyond subjective and objective in statistics (joint work with Andrew Gelman). Tilburg, June 2015
- Model assumptions and
truth in statistics. UCL Science and Technology Studies, February 2015
-
If all models are wrong , are the robust ones better?
The role of robust statistics in a constructivist philosophy of
statistics. St. Petersburg, July 2013
- Measurement as a
constructive act - a statistician's view. Bielefeld, March 2013.
- How statisticians quantify
evidence. UCL, March 2012 (in the Graduate School series on
interdisciplinary studies of evidence)
- Statistical models,
estimation and reality. UCL, February 2012 (introduction to some
statistical concepts for the Energy Institute)
Tutorials
Other still unpublished material
- Decisions that are needed when using cluster analysis, and research that helps with making them. Plenary keynote lecture, IFCS, Tokyo, August 2017.
- Gaussian and not-so-Gaussian clustering with robustness against outliers and a stab at the number of clusters (joint work with Pietro Coretto). Model based clustering and classification, Catania, September 2016.
- Exploration of the variability of variable selection based on distances between bootstrap sample results (joint work with Willi Sauerbrei). DagStat, Göttingen, March 2016.
- Cluster validity indexes: calibration, aggregation, further issues. ERCIM, London, December 2015.
- Football and the dark side of cluster analysis (poster, joint work with Serhat Akhanli). Milton Keynes, May 2015.
- Discussion of the invited
session on the number of clusters. Tilburg, July 2013.
- The aggregation of variables in
distance design - how to get more out of distance-based methods.
London, December 2010.
- Discussion of the
invited session on data visualisation. Florence, September 2010.
- Some thoughts about robust
clustering. Aachen, October 2009.