News for hackers. Stuff that matters.
Re:July 2015 was warmest month ever recorded by fluidcruft on Thursday August 20, 2015 @03:45PM (#10094167)
You've choesn to insist on replacing the actual continuous data with a weaker derived parameter (period between record breaking events). Of course tests on weakend data have lower power. Should I be surprised that a test which relies on data extremes has less trend detection power?

But I'll also note that you haven't even bothered to go back and compute the probabilities in the first place.

It was one thing to make this argument in the 90's, but two decades of data later it's plainly ridiculous and it doesn't take nuanced data analysis.

https://www.ncdc.noaa.gov/sotc/service/global/glob/201506.gi...

https://www.ncdc.noaa.gov/sotc/service/global/lo-hem/201506....

There are no comments on this comment yet.