Friday, February 08, 2013

Open for Comment

Steve Bloom points to Comment on “Polynomial cointegration tests of anthropogenic impact on global warming by Beenstock et al (2012)".  Some fallacies in econometric modelling of climate change" D. F. Hendry and F. Pretis.

Now Eli has to admit a bit of shame about this one.  The bunny got tangled up in the mathturbation.  James, in his usual laconic way was the closest

Were I Judith Curry, I would probably be saying "wow" at this stage. Alternatively, it could just be some dross that has accidentally found its way into print after having been rejected at least twice at different journals.
The review comments are interesting, to say the least. Reviewer #2, in particular, seems awfully keen on a number of silly sceptic claims that have been presented in recent years.

I suppose it just goes to show that you can fool at least one person sometimes, and if that person happens to be a journal editor, you're in luck.
Those of you who remember Eli and Socrates going around on statistics know the answer
 [Eli] In other words, if you have a good idea of the answer they can help you, but if not you need physics or biology or chemistry or meteorology.
Hendry and Pretis, smart bunnies, didn't look at the econometric analysis, they looked at the data set.  Why you ask, well, anyone who studies blog scientists knows why
In their analysis of temperature and greenhouse gases, Beenstock et al. (2012) present statistical tests that purport to show that those two variables have different integrability properties, and hence cannot be related. The physics of greenhouse gases are well understood, and date from insights in the late 19th century by Arrhenius (1896). He showed that atmospheric temperature change was proportional to the logarithmic change in CO2). Heat enters the Earth’s atmosphere as radiation from the sun, and is re-radiated from the warmed surface to the atmosphere, where greenhouse gases absorb some of that heat. This heat is re-radiated, so some radiation is directed back towards the Earth’s surface. Thus, greater concentrations of greenhouse gases increase the amount of absorption and hence re-radiation. To “establish” otherwise merely prompts the question “where are the errors in the Beenstock et al. analysis?”.
In other words, the Beenstock et al don't know anything about the system they are studying.  In particular, point out that the data series used for greenhouse gas concentrations, is not a single series but a compilation, and that the nature of the data changes over about 1960 from ice cores to atmospheric grab samples
Interacting with unmodelled shifts, measurement errors can can lead to false interpretations of the stationarity properties of data. In the presence of these different measurements and structural changes, a unit-root test on the entire sample could easily not reject the null hypothesis of I(2) even when the data are clearly I(1). Indeed, once we control for these changes, our results (see Tables 1 and 2 below) contradict the findings in Beenstock et al. (2012)


 Once that is done, and one actually looks at the data, it becomes clear that there are two separate  periods during which the properties of the correlation between temperature and forcing changes, roughly divided at 1960, and that Beenstock's analysis depends on an incorrect pooling of data.

Econometrics is a hammer which econometricians apply to all objects, but  it's ALWAYS the science that rules.

46 comments:

Rattus Norvegicus said...

Beenstock, et. al. have been trying to get some variant of that published for years. It looks like ol' Jack has run into the giant this time and the giant has figured out the grift...

James Annan said...

BTW, the editor who handled the Beenstock paper has resigned.

Steve Bloom said...

Ah, must have been my suggestion. :)

More seriously, perhaps this is reading too much into her comment approving publication, but she sounded like she was either overextended or under stress of some sort.

So now, does this paper meet the standards for a retraction?

Anonymous said...

Steve: the answer to your last question is "no", especially if this particular comment paper goes through. In general then the scientific community considers it sufficient. Of course, the *authors* may still decide to retract the paper.

Anonymous said...

"Fee, Fi, Fo-Fum, I smell the turd
of an econometrician"

-- by Horatio Algeranon

The Beenstock caused
Some big sensations
When it grew
To Giant stations

Blasted out
The Greenhouse top
Where fate had caused
The seeds to drop

Grew to Econo-Land
In the sky
Where polynomials
Often lie

Where Mathgic geese
Are golden birds
With golden eggs
And golden turds

Where Econometricks
Rules the day
And mathturbation
Holds the sway

The Giant really
Didn't know jack
Which gave the ax
To mathgic "facts"

From gravity
The Beenstock fell
For physics rules
As scientists tell.

The End

Anonymous said...

Econometrics is a hammer which econometricians apply to all objects, but it's ALWAYS the science that rules.

Indeed. Now go tell that to Tamino, who "calculates" an enormous, magically-fast effect of solar forcings by dumb regression on recent data (bonus: it changes drastically if you use older data). Because some people have a problem with this.

(Not that it would necessarily change the overall conclusions, but the parallel was just too amusing.)

J Bowers said...

"The temperature rise from 1820 to 1950 can be attributed for about 70 % to increased solar irradiance, while the temperature changes after 1950 are almost completely produced by the interplay of anthropogenic greenhouse gases and aerosols."

I'm sure some people will have a problem with that.

Anonymous said...

Eli, VS, who is your professional superior, considered you a clown in 2010. By this post I guess that he still does.

EliRabett said...

No, what she is is an land use expert, much like Roger Sr. and there is a lot of sense to that for the period 1850-1920/30 when the North American west and Australia and parts of Africa were cleared for farms. AFAEIC anyone who says it was all solar in those periods had better account for the pioneer effect (which they don't).

willard said...

For what it's worth, dear frolicsome bunny, VS' trick was a variant of this old one:

http://www.amazon.com/tag/science/forum/ref=cm_cd_pg_pg16?_encoding=UTF8&cdPage=16&cdSort=oldest&cdThread=Tx3TXP04WUSD4R1

VS' trick lasted until a guy in machine learning showed up, btw.

Steve Bloom said...

Re the link discussed by anon and JB above, are fractals the new epicycles? Seems like they do the job.

Eli, since when was RP Sr. a land use expert? Back in the day (when he had a blog that allowed comments) he seemed pretty ignorant of the subject. E.g. I recall that when he first got on to the subject he made a broad assertion that the IPCC had largely ignored land use in favor of other anthropogenic forcings, a notion which was falsified by looking at the first page of the AR4 Synthesis. He seemed quite then on the details of the relevant science as well.

Horato, that's brilliant! You're only improving with time.


Steve Bloom said...

then => thin

David B. Benson said...

willard --- Vaughan Pratt is a computer scientist but I would hardly characterize him as a guy in machine learning.

Anonymous said...

No the late editor _eli

willard said...

David,

I was referring to Igor Samoylenko:

http://ourchangingclimate.wordpress.com/2010/03/01/global-average-temperature-increase-giss-hadcru-and-ncdc-compared/#comment-3059

My memory was that he was into machine learning, but it can fail me. I can't find his research page anymore.

David B. Benson said...

There is a Igor Samoylenko @ Univ. Bath:
http://www.opengrey.eu/item/display/10068/697475

Anonymous said...

I will tell ya, that blog(Bart V) some 2 1/2 years ago was an intellectual tour de force. One man, VS, against an avalanche of statistically-challenged climate scientists was a site to behold. Might even go down as one of the greatest smack-downs in history. Eli declared a clown, Bart mumbling to himself, Tamino proved to be a "fake" statistician, Zorita corrected on the spot and apologies for the mistakes in his recent paper(so much for climate peer review),and Annan so frightened he didn't even enter the fray. Yet Annan show up here as a gossiper - Oh, I heard the reviewer of Beenstock resigned. What a mann!

Steve Bloom said...

Those Galt wannabes are something to behold. Imagine what an ego boost it must be to be able to restrain AGW through sheer force of will.

J Bowers said...

If anyone's wondering who VS is and how he owned Bart's blog, you can judge for yourselves:

* Global average temperature increase GISS HadCRU and NCDC compared
* Is the increase in global average temperature just a ‘random walk’?

metzomagic said...

It appears that Anonymous above has a different view of reality than most of us here. How can every recent decade being hotter than the last and ever-increasing ocean heat content (and ever-decreasing arctic ice) be attributed to a 'random walk'?

OK, so VS's mathturbation skilz might be a cut above the usual variety as seen on WUWT, but it's still just that. You can't massage data in the absence of understanding the underlying physical processes at work, as VS and Steve McIntyre so ably demonstrate... with the D-K challenged likes of Anonymous cheering them on.

willard said...

> Zorita corrected on the spot and apologies for the mistakes in his recent paper.

A quote would be nice.

Here's one:

> With all respect, to keep this comment short, it seems that you did not read that GRL paper, and perhaps no one of the commenters here did. That GRL paper was not about ‘random walk’ or ‘unit root’ whatsoever. We tested the H0 hypothesis that the global annual mean temperature could be represented by a non-deterministic fractional-differenced process- not a random walk- and thus stationary.

http://ourchangingclimate.wordpress.com/2010/03/01/global-average-temperature-increase-giss-hadcru-and-ncdc-compared/#comment-2068

Anonymous said...

metzomagic, for some reason climate scientists always bring up straw men arguments, like it is the physics that matters, not the stats. Sure physics matters but not for the purposes being discussed. It matters not a wit where the data comes from, whether from physics or grocery sales numbers. It is how you treat the data and form observations and conclusions using state-of-the art statistical reasoning that really matters. Climate prophets, like Eli, really need to learn and accept this or no one is going to take them seriously.

J Bowers said...

Doing the maths without doing the science is not doing the science.

Anonymous said...

J Bowers, you repeatedly miss the point. No one says don't do the science. It is how you treat the data once the science is done that really matters. It is truly amazing that needs to be pointed out.

Willard, it was a 30 something day blog.. Go read it, it is there.

dhogaza said...

"metzomagic, for some reason climate scientists always bring up straw men arguments, like it is the physics that matters"

This needs to be posted on a billboard somewhere.

The fun thing about VS's work is that it leads to a statistical model predicting that in the next 50 years, the rise in global temps will be 0 +/- 1C. Which essentially makes it unfalsifiable in that timeframe ...
even a statistically significant rise (tested by ignoring his unit root claim) over that period of time lies within his "drunken walk" prediction so ... victory! Stats trumps physics!


J Bowers said...

"Some people use statistics as a drunken man uses lamp-posts for support rather than illumination."
source

Anonymous said...

dhogaza, have you re-read your postings in that 2010 blog. I would not boast about anything If I were you. You were smacked silly. And don't quote out of context. We say over and over, the physics is important, but the conclusions you draw from the data out of that physics is what is important. How thick are you. Mann, I hope you are not in a teaching role.

Saeed Zia said...

Jobs at your Home, Internet Online Jobs like data entry, copy pasting, Form Filling, Facebook Sharing Jobs, Clicking Jobs, Web Surfing, Google Jobs and Much More Earning Systems Online
www.jobzcorner.com

J Bowers said...

* Not a Random Walk
* Still not

Still not seeing VS smacking down anyone, just a lot of deniers at Bart's saying he did. Some people are easily pleased.

dhogaza said...

So I'm sure anon can show us where VS published his astonishing statistical analysis which demolishes everything known about climate science.


Such an important result would have journal editors fawning over each other for the privilege of publishing it.

Or was it just a case of blog mathurbation?

You see, anon, it is easy to claim that all we know about climate physics is bunk. It is much harder to actually establish this as fact.

2010 ... it is now 2013 ... and oddly enough, physics-based climate science continues to march forward, unhindered by econometrics.

Anonymous said...

J Bowers, if you had bothered to read he did not, I repeat not suggest a random walk. Only those incapable of understanding the discussion thought so.

As for you dhogaza, I understand how you might feel. You and sod have a very rough time hanging in there. The Mannly thing to do would to admit you were thumped. Now, off with you

a_ray_in_dilbert_space said...

dhogaza: "2010 ... it is now 2013 ... and oddly enough, physics-based climate science continues to march forward, unhindered by econometrics."

THIS.

The problem is that the econometricians fundamentally do not understand that the goal of science is prediction, not mere explanation. So, meanwhile climate science marches on while all the denialists can manage is: "Oh, woe is me. It's all too complicated. It's all too uncertain!"

Science works. Econometrics...[looks back at the 2009 collapse, sovereign debt and Pets.com]...not so much.

EliRabett said...

Whatever>

Anonymous said...

a_ray_in_dilbert_space said.
"The problem is that the econometricians fundamentally do not understand that the goal of science is prediction, not mere explanation. So, meanwhile climate science marches on while all the denialists can manage is: "Oh, woe is me. It's all too complicated. It's all too uncertain!"

Science works. Econometrics...[looks back at the 2009 collapse, sovereign debt and Pets.com]...not so much

You guys really are stuck on stupid. Statistics tells you about the probability of your experiment's brush with the null hypothesis, not to predict the future. My, oh my!

willard said...

Lazy bunny,

I know it's a long thread, but it's your claim, so it's for you to back this up. You can search for "eduardo Says" to speed things up.

Here's another quote from Zorita:

> [T]he recent clustering of record annual temperatures is very unlikely in a long-term-persistence process (fractional differencing process), and therefore points to either a non-stationary or deterministic trend in the period analyzed. This statistical analysis cannot discriminate between causes, as only temperature data were analyzed , not forcing data.

http://ourchangingclimate.wordpress.com/2010/03/01/global-average-temperature-increase-giss-hadcru-and-ncdc-compared/#comment-2079

Would this count as evidence of what you're suggesting?

You might prefer to find Eduardo's quote yourself.

Steve Bloom said...

dhogaza: "Such an important result would have journal editors fawning over each other for the privilege of publishing it."

May I suggest instead:

"Such an important result would have journal editors resigning after having made the mistake of publishing it."

That seems to happen rather frequently with climate science papers from economists and statisticians, although sometimes the editors are merely badly embarrassed.

As pointed out above VS never even submitted anything to a journal. That's OK, though, as these Galtian ubermensch don't need no stinking journals to bend reality to their will.

a_ray_in_dilbert_space said...

Anon,
Just about completely wrong. Stats tells you the significance of your results IF you have the right model. If not, you will simply fool yourself.

As to the rest of your silliness:

1)The null is simply a tool--it is necessary because statistical results are comparative rather than absolute.

2)If prediction is unimportant then fit your 6 data points with a 6-parameter polynomial and declare victory.

John Mashey said...

a_ray:
Your comment on econometrics is incorrect: that had nothing to do with pets.com.

Anonymous said...

The Quants
-- by Horatio Algeranon

They're on a random walk
Down Wall Street, with a bottle.
The Quants, they love to talk,
But empty is their model.*

*Their bottle too.


Anonymous said...



VS and the big brown Crayon
-- by Horatio Algeranon

VS has a big Brown crayon
That leaves a Brownian track
He's on a trek to Randumb Land
And never will come back.

Anonymous said...

If Horatio takes requests, I suggest "Comfortably Dumb".

Rib Smokin' Bunny

J Bowers said...

"he did not, I repeat not suggest a random walk"

He did:

VS Says: March 4, 2010 at 13:54
"In other words, global temperature contains a stochastic rather than deterministic trend, and is statistically speaking, a random walk."

Anonymous said...

I recall the VS exchanges and in my recollection (s)he was left black and blue, anally speaking.

And, what's more, the "random walk" nonsense has never raised its head again.

Man with no Name

Hank Roberts said...

> econometrics ... pets.com

Assessing "How much is that doggie in the window?" predates even the Internet.

J Bowers said...

Cue sound effect

Anonymous said...

"Flat as a Pancake"
-- by Horatio Algeranon (with some help from Aunt Jemima)

Temperature statistics
Without the physics
Is like the spring
Without the fall
There's only one thing worse
In this universe
That's not any science at all.