Discussion Papers, Department of Business and Management Science, Norwegian School of Economics (NHH)
Best estimate reporting with asymmetric loss
() and Richard Sinding-Larsen
Abstract: This paper considers the problem of point prediction based
on a predictive distribution, representing the uncertainty about the
outcome. The issue explored is the reporting of a single characteristic,
typically the mean, the median or the mode, in the context of a skewed
distribution and asymmetric loss. Special attention is given to the
two-piece normal distribution and asymmetric piecewise linear and quadratic
loss. The practical context for the issue is the yearly reporting of
remaining petroleum resources given by the authorities to stakeholders that
may ask for just a single number.
Keywords: Estimate reporting; asymmetric loss; point prediction; predictive distribution; (follow links to similar papers)
JEL-Codes: C00; C10; C13; (follow links to similar papers)
15 pages, January 30, 2015
Before downloading any of the electronic versions below
you should read our statement on
for viewing Postscript files and the
Acrobat Reader for viewing and printing pdf files.
Full text versions of the paper:
Questions (including download problems) about the papers in this series should be directed to Stein Fossen ()
Report other problems with accessing this service to Sune Karlsson ()
or Helena Lundin ().
Design by Joachim Ekebom