It is time to stop quivering in our boots in pointless fear of the future and just roll up our sleeves and build it.
- Ray Pierrehumbert

Thursday, July 16, 2009

The Return of Forecasting Principles


Remember "Forecasting Principles"? 

J. Scott Armstrong, a B-school investment forecaster, took on climate theory based on the validity of financial forecasts! Sure enough, he concluded that we are overextended

Not satisfied with the fool he made of himself at the time, Armstrong was quick to take up the invitation (!) from the Financial Post to revisit the question in the light of the MIT probabilistic forecast. (I am inclined to be skeptical, myself, about that publication, but that doesn't mean I am willing to make up silly counterarguments.)

Armstrong's substantive contribution, in association with Dr. Willie Soon (of the "Harvard Smithsonian Center for Astrophysics" no less) is as follows:
With Dr. Willie Soon of the Harvard-Smithsonian Center for Astrophysics, we found that simply predicting that global mean temperatures will not change results in quite small forecast errors. In our validation study that covered the period 1851 to 2007, we compared the no-change forecast with the IPCC global warming forecast that temperatures will climb at a rate of 0.03C per year. We compared the IPCC projection of 0.03C per year with what actually happened after 1850. The errors from the IPCC projection were 12 times larger than no-change benchmark. Consider the accuracy of the no-change model: On average the 50-years ahead forecasts differed by only 0.24C from the global mean temperature as measured by the Hadley Centre in the U. K.
What a clever test! It turns out that the temperature change over the past 150 years is closer to zero than to 4.5 C. Therefore, wait for it...
Based on our analysis, we expect the annual global mean temperature for every year for the rest of the 21st Century to be within plus-or-minus 0.5C of the 2008 mean.
Therefore the sensitivity is zero! Haha! Look, the three lines of code were written by an astrophysicist! Zero! Hahahaha!

8 comments:

Aaron said...

I like the way he was able to forecast the 2008 credit crunch/ recession. Oh! Right! he did not. Never mind. And polar bears, he took Fish and Game to task for using their peer reviewed forecasting method developed by a federal rule making process and required by regulation. (Which seems to have done a better job of forecasting bear populations than Armstrong’s.)see (http://econpapers.repec.org/paper/pramprapa/6317.htm)

Tom Fiddaman said...

These guys are getting dumber. They're backcasting to 1850 at 3C/century! From http://kestencgreen.com/naiveclimate.pdf :

"It is not unreasonable, then, to suppose for the purposes of our validation illustration that scientists in 1850 had noticed that the increasing industrialization of the world was resulting in exponential growth in “greenhouse gases” and to project that this would lead to global warming of 0.03°C per year."

Not unreasonable, if you have no concept of the relationship between GHG concentrations and radiative forcing or stock-flow dynamics of heat!

My takedown of the old Armstrong/Green stuff is here http://blog.metasd.com/?s=armstrong , but this will be an even easier target.

Marion Delgado said...

Michael, his efforts would be a-dime-a-dozen, if it were not for the fact that he pretends to be some sort of institute that publishes journals, and so on.

He reminds me strongly of the Idsos, in fact.

Tom Fiddaman said...

Armstrong & Green are contributors to the Idso/Singer NIPCC report, so some similarity is not surprising.

amoeba said...

I can't think of anything polite or non-venomous to say regarding these ahem individuals.

Straw man indeed!

AARRGH! Head explodes with rage!

James Annan said...

It is established beyond reasonable doubt that Armstrong's ideas aren't worth the electrons to discuss them. However, I've seen several of your asides in the direction of the MIT thing and would be interested in a lengthier exposition...

James

Tom Fiddaman said...

A bit more detailed look here
http://blog.metasd.com/2009/07/17/unprincipled-forecast-evaluation/

Marion Delgado said...

WHAT IS THE LAW?

A SCIENTIST PUBLISHES IN A JOURNAL

THAT IS THE LAW!

WHAT IS THE LAW?

A SCIENTIST MEASURES THINGS!

THAT IS THE LAW!

FOR ARE WE NOT SCIENCE??

From The Island of Dr. Armstrong