@DrEricDing @edgarrmcgregor @ClimateHuman Utterly flawed approach to (1) assume normal distribution and (2) ignore serial correlation in climate statistics. Here's a paper where a 10^-17 p-value completely disappears to insignificance once these are accou
579 followers
579 followers
@Large_Eddy @RARohde Care needed in gaussian/independent assumptions, or that 10^-12 means assumptions don't matter. Reminds me of this paper when statistical likelihood of an event went from 10^-17 to 0.07 by modifying the assumptions of the underlying s
579 followers
@ed_hawkins @DavidTCDavies @metoffice Full paper and reference here: https://t.co/HK5ynGHj3Q
604 followers
Papers on my deep dive reading list. 1. https://t.co/MOaEC9W3MG