Dikran Marsupial’s excellent bit on hypothesis testing applied to climate, or how it should be applied, if at all

Frankly, I wish some geophysicists and climate scientists wrote more as if they thoroughly understood this, let alone deniers to try to discredit climate disruption. See “What does statistically significant actually mean?”.

Of course, while statistical power of a test is important to keep in mind, as well as the effects of arbitrary alterations or recodings of data upon it (see also Andrew Gelman’s comment on this), people should really look at this from a purely Bayesian perspective, and there’s no longer a computational excuse to ignore that approach.

About hypergeometric

See http://www.linkedin.com/in/deepdevelopment/ and https://667-per-cm.net/about
This entry was posted in Anthropocene, anti-science, Bayesian, climate change, climate data, climate disruption, D. K. Marsupial, Frequentist, global warming, hiatus, Hyper Anthropocene, ignorance, John Kruschke, regime shifts, statistics, Student t distribution. Bookmark the permalink.