Local and discrete scoring rules

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

• Matt Parry (Cambridge)
• Friday 14 March 2008, 14:00-15:00
• MR12, CMS, Wilberforce Road, Cambridge, CB3 0WB.

If you have a question about this talk, please contact R.B.Gramacy.

At its simplest, a scoring rule is a loss function for choosing a certain distribution to represent the uncertainty of a random variable. (A scoring rule is not a score function!) One of the key requirements is honesty: the expected score must be minmised by choosing the true distribution—- if it is known. I take an introductory tour of the role of scoring rules in decision theory and statistical inference and of their connection to entropy and geometry. I end with a discussion of recent advances in the use of scoring rules in cases of both continuous and discrete sample spaces.

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.