Lies, damnable lies and statistics.
It all depends on your choice of models. Statistical technicques can only tell if the observations fit the model. Anthropogenic Global Warming Theory is very robust with respect to data, all observations confirm it at the 100% confidence level. The great success of the AGWT advocates is the care with which they select their observations and the flexibility they show in patching their model(s). (Do. I. really. need. sarcasm. tags. here?)
I have no argument with these statisticians of the AP about the last ten years.
Anyone who wants to go through my posts over the last 1-2 years will see that I have been advising caution about the “cooling” trend.
HOWEVER, I also do not believe that there is any strong evidnence for a statistically significant warming trend over the last 130 years, and more especially that the minor warming trend that has occured is significantly correlated with CO2.
By the way, I believe that any serious scientist now must eliminate the HadCrut Data from their purview, and only use the US Temperatures, and the Satellite data. (Historical data including tree ring data must also be considered corrupt). Other world wide data has been corrupted beyond repair.
In my political methods class, I learned that statistics can be used to lie about a subject.