Showing posts with label significance. Show all posts
Showing posts with label significance. Show all posts

Saturday, July 14, 2012

"Statistically significant scenarios"

Sometimes, scenarios of possible future climate change are examined if they would represent a "statistically significant" change from present conditions. Usage of this terminology is misleading, and valid only in a very restricted sense - because of the sampling assumptions needed for employing the machinery of statistical hypothesis testing.
This issue is discussed in the paper  "Testing ensembles of climate change scenarios for"statistical significance"by Hans von Storch and Francis W. Zwiers, which has been accepted for publication by "Climatic Change".