Choosing a good histogram

Add to your list(s) Download to your calendar using vCal

• Yannick Baraud (Univ. Nice)
• Friday 16 October 2009, 16:00-17:00
• MR12, CMS, Wilberforce Road, Cambridge, CB3 0WB.

If you have a question about this talk, please contact Richard Nickl.

Histograms are probably among the most simple and popular estimators. They are widely used in science, especially by non-statisticians, in view of estimating densities (or intensities of point processes). To build a good histogram, one needs to partition the data in a suitable way which turns out to be a tricky problem. Given a (possibly large) family of candidate partitions, how can we select a suitable one on which our histogram will be as close as possible to the unknown density? Besides, are there families of partitions one should consider preferably? These are some of the questions we shall try to answer in this talk by adopting a non-asymptotic point of view.

This talk is part of the Statistics series.

This talk is included in these lists:

• All CMS events
• All Talks (aka the CURE list)
• CMS Events
• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge talks
• Chris Davis' list
• DPMMS Lists
• DPMMS info aggregator
• DPMMS lists
• Guy Emerson's list
• Hanchen DaDaDash
• Interested Talks
• MR12, CMS, Wilberforce Road, Cambridge, CB3 0WB
• Machine Learning
• School of Physical Sciences
• Statistical Laboratory info aggregator
• Statistics
• Statistics Group
• bld31
• custom
• rp587

Note that ex-directory lists are not shown.