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

Learn More →

Reply to “Comments on ‘H. L. Wagner’s Unbiased Hit Rate and the Assessment of Categorical Forecasting Accuracy’”

Reply to “Comments on ‘H. L. Wagner’s Unbiased Hit Rate and the Assessment of Categorical... 1. Introduction By way of cautioning atmospheric scientists, Jolliffe and Stephenson identify six purportedly undesirable properties of Wagner’s measure, H u , which I failed to discuss in Armistead (2013) . The authors do not question the two major thrusts of the paper: that H u avoids limitations that variously affect single-expression categorical measures in the atmospheric sciences, including those proposed by Doolittle, Gilbert, Heidke, Peirce, and (more recently) Stephenson, and that there is a logical linkage between Wagner’s measure, Bayes’s theorem, and Murphy and Winkler (1987) . Except where necessary to clarify a point, my discussion is limited accordingly. In brief, this reply argues as follows. Two of the authors’ observations do not identify properties of Wagner’s measure, whether undesirable or otherwise; instead, they reflect what could be considered advantages of the properties of H u ( sections 2a and 2b ). A third observation ( section 2c ) reflects an apparent misunderstanding of a property of H u , and three observations ( sections 2d – f ) are mixtures of well-considered opinions with which I do not agree and of further possible misunderstandings. I believe that none of the six observations rises to the level http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Weather and Forecasting American Meteorological Society

Reply to “Comments on ‘H. L. Wagner’s Unbiased Hit Rate and the Assessment of Categorical Forecasting Accuracy’”

Weather and Forecasting , Volume 29 (3) – Jan 14, 2014

Loading next page...
 
/lp/american-meteorological-society/reply-to-comments-on-h-l-wagner-s-unbiased-hit-rate-and-the-assessment-x9FmGETxmg

References (4)

Publisher
American Meteorological Society
Copyright
Copyright © 2014 American Meteorological Society
ISSN
0882-8156
eISSN
1520-0434
DOI
10.1175/WAF-D-14-00008.1
Publisher site
See Article on Publisher Site

Abstract

1. Introduction By way of cautioning atmospheric scientists, Jolliffe and Stephenson identify six purportedly undesirable properties of Wagner’s measure, H u , which I failed to discuss in Armistead (2013) . The authors do not question the two major thrusts of the paper: that H u avoids limitations that variously affect single-expression categorical measures in the atmospheric sciences, including those proposed by Doolittle, Gilbert, Heidke, Peirce, and (more recently) Stephenson, and that there is a logical linkage between Wagner’s measure, Bayes’s theorem, and Murphy and Winkler (1987) . Except where necessary to clarify a point, my discussion is limited accordingly. In brief, this reply argues as follows. Two of the authors’ observations do not identify properties of Wagner’s measure, whether undesirable or otherwise; instead, they reflect what could be considered advantages of the properties of H u ( sections 2a and 2b ). A third observation ( section 2c ) reflects an apparent misunderstanding of a property of H u , and three observations ( sections 2d – f ) are mixtures of well-considered opinions with which I do not agree and of further possible misunderstandings. I believe that none of the six observations rises to the level

Journal

Weather and ForecastingAmerican Meteorological Society

Published: Jan 14, 2014

There are no references for this article.