Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

letters to the editor1

letters to the editor1 letters to the editor' Forecasting Skill of Beginners On the other hand, it is possible in the case of probabilistic forecasts to formulate proper scoring systems, and the squared I would like to make some comments concerning several facets error rule used by Gedzelman is proper. In fact, this rule is of the recent paper by Gedzelman (1978) on the performance equivalent to the well-known Brier score (Brier, 1950). If of forecasters in the CCNY forecasting contest. forecasters attempt to minimize their (subjective) expected First, I agree with Gedzelman that new forecasters gen- scores, then such a system will discourage hedging, not en- erally attain a level of "skill" similar to that of experienced courage it as stated by Gedzelman. With proper scoring sys- forecasters in a relatively short period of time. Reliability in tems (and under the stated assumptions), it is impossible to probability forecasting provides another example of this rapid beat these systems by hedging. Thus, the scoring rules used improvement in performance. Initially, forecasters frequently by Gedzelman necessarily encourage hedging in the case of assess probabilities that are too extreme and quite unreliable. categorical forecasts, but they discourage it in the case of However, the results http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bulletin of the American Meteorological Society American Meteorological Society

letters to the editor1

Bulletin of the American Meteorological Society , Volume 60 (6): 3 – Jun 1, 1979
Download PDF
3 pages

Loading next page...
 
/lp/american-meteorological-society/letters-to-the-editor1-THtMlJ0SGx
Publisher
American Meteorological Society
Copyright
Copyright © American Meteorological Society
ISSN
1520-0477
eISSN
1520-0477
DOI
10.1175/1520-0477-60.6.691
Publisher site
See Article on Publisher Site

Abstract

letters to the editor' Forecasting Skill of Beginners On the other hand, it is possible in the case of probabilistic forecasts to formulate proper scoring systems, and the squared I would like to make some comments concerning several facets error rule used by Gedzelman is proper. In fact, this rule is of the recent paper by Gedzelman (1978) on the performance equivalent to the well-known Brier score (Brier, 1950). If of forecasters in the CCNY forecasting contest. forecasters attempt to minimize their (subjective) expected First, I agree with Gedzelman that new forecasters gen- scores, then such a system will discourage hedging, not en- erally attain a level of "skill" similar to that of experienced courage it as stated by Gedzelman. With proper scoring sys- forecasters in a relatively short period of time. Reliability in tems (and under the stated assumptions), it is impossible to probability forecasting provides another example of this rapid beat these systems by hedging. Thus, the scoring rules used improvement in performance. Initially, forecasters frequently by Gedzelman necessarily encourage hedging in the case of assess probabilities that are too extreme and quite unreliable. categorical forecasts, but they discourage it in the case of However, the results

Journal

Bulletin of the American Meteorological Society – American Meteorological Society

Published: Jun 1, 1979

There are no references for this article.