Sheldon Walker decided to write a post suggesting “A possible compromise on global warming slowdowns and pauses.” He even offers the conciliatory admission “So we are effectively arguing about different things. This means that we could both be right (or we could both be wrong).”
Here’s what his “compromise” sounds like to me:
Or maybe his argument is like the Norwegian Blue parrot–“He’s not dead, he could be pining for the fjords…”
yep, that could have gone either way
“I don’t know about you, but I am getting a bit fed up with the warmist lie, that deniers claim that a slowdown or pause started in 1998.”
It’s not very difficult to find such refrains (apologies for all the WUWT links):
(^in the comments it’s such a common refrain)
Monckton tried gaslighting people by starting in 1997, ostensibly solving the issue of the 1998 El Nino being at the beginning, but all of his posts fall victim still to the same problem.
Walker COULD have actually been born a couple Thursdays ago, can’t rule that out.
[Response: There’s an even more recent example, discussed here.]
Perfect video illustration that saved us from reading thousands of words of argumentation. I just listened to a talk produced by John Cook that discusses this kind of argumentation:
Ep 20: Dave Roberts on conservatives and climate change
and about the intractability of conservatives on climate change and whether polarization is something to be avoided or embraced.
Listen to the episode at http://evidencesquared.com/ep20/
Most folk probably have little time for reading turgid crap written by Sheldon Walker. Me, I enjoy a good laugh so can share with you what Walker thinks he’s about with his latest Wattsupian writings that has led to talk of “compromise” ha ha ha.
Walker says he was annoyed by a recent post of our host Tamino here at Open Mind (concerning the statistical relevance of short slowdowns, something beloved by Walker but which specifically makes not reference to Walker’s bullshitting). Walker’s annoyance was such that he “started to write an article to document his [Tamino’s] many errors.” He apparently got halfway through this writing before deciding he should take on the role of the Monty Python legless armless bity knight.
(But having turd for brains, Walker still nailed his grand half-a-rebutal onto the boghouse door at Wattsupia, 1,200 words that I defy anyone to identify within it even one useful assertion that stacks up.)
The basic position Walker then sets out is that he does not deny the continued existence of a signal from AGW. Rather he tells us:-
He thus ignores the point that his grand analysis of temperature anomalies lacks statistical significance relative to the AGW signal and thus that he demonstrates no pause. (Strangely, identifying a period within a temperature record which has a statistically significant pause is not an impossible task. But it does remain a silly waste of time.)
And having declared that denialists (skeptics) are like him and are mainly closet AGW-believers (but with th caviat ” I cannot speak for all skeptics”), Walker then pretty-much ignores the evidence that strongly suggests that if he were truly of such a mind he would be the only one like it on the planet Wattsupia. His claim that the ‘pause’ is all a big misunderstanding is shown up for what it is by the comment thread set out on the bogroll under his Wattsupian OP.
The first three primary comments run as follows:-
So it looks like Sheldon Walker is both wrong and friendless; delusional and incompetent.
“At times, the various influences, especially the pseudo-random element, make the global temperature series appear to slow down or pause. This is what the skeptics are talking about when they claim that there is a slowdown or pause over a certain timeframe”
Or, as Bill O’Reilly would say:
“Temps go up, temps go down. You can’t explain that.”
I got as far as this comment: “I thought all Sheldon’s are geniuses (sarc). Why waste your time reading Foster’s trash when you know its invalid and is only going to piss you off. Start concentrating on the impending Constitutional Crisis if the hard left players don’t go to jail.”
I could read no further.
I think Wattsupians voted heavily for Trump.
Sheldon Walker: Oh, oh, I see, running away then. You yellow
bastards! Come back here and take what’s coming to you.
I’ll bite your legs off!
A quick update on some comments that you made.
1) That I would not be able to learn how to do a linear regression with a correction for autocorrelation, off the internet.
– I will give you half marks for this comment. I found most of the stuff on the internet about autocorrelation too technical. I wanted a practical example. So I had to work it out for myself. I triple checked my results to make sure that they were right. For each date interval I did 3 linear regressions:
1) first regression with no correction for autocorrelation. Used this to compare to the other regressions, to make sure that values were reasonable.
2) added lagged temperature anomaly. Did linear regression to work out how much autocorrelation was present. Over 741 regressions the average was about 0.58 (that figure is from memory, may be wrong). The amount of autocorrelation varied from about 0.4 to 0.7.
3) corrected the temperature anomaly for autocorrelation. Did another linear regression to make sure that it was gone. All 741 regressions had a residual autocorrelation of about 1.0 x e^-15
2) That I would not learn how to do a linear regression with a correction for autocorrelation overnight.
– You get full marks for this comment. It took me 6 days. Considering that I have a full time day job, I think that 6 days is ok.
I have fully analysed the GISTEMP data using the method that I developed. I found 9 slowdown trends which are significant at the 99% confidence level. They mostly start in 2001 and 2002. The longest lasts 14 or 15 years (sorry, can’t remember which – I don’t have the results with me). There were also 2 slowdown trends which are significant at the 95% confidence level, and 2 or 3 which were significant at the 90% confidence level.
Interestingly, there were 6 slowdown trends which were significant at the 90% confidence level, which started in 1997 and 1998. The famous slowdown that warmists say is due to the 1998 El Nino. They are much less significant then the 2002 slowdown.
Watch out for my results on WattsUpWithThat. It will be a while, because I want to include the same analysis for UAH and RSS. I have not done those yet.
I offered warmists a compromise about slowdowns. You turned it down. You won’t be offered it a second time.
All the best,
[Response: See my latest post.]
But for all that, do you really think that any “slowdowns” are really more than just noise? Be honest. Because the simplest way I can think of when describing the data is a steady trend with noise. Sure you could come up with more complex models, but why would you do that?
Sheldon, we are talking about explanation of what is actually happening, This is not a matter of opinion, either there has been a change in trend large enough to matter or there has not. One might offer a compromise on a matter of policy. But one does not negotiate an explanation. One seeks as clear, and comprehensive an explanation ion as one can come up with. Yes one listens to critics if one believed that the criticism was given in good faith and one might modify ones opinions as a result. That is not negotiating or compromising or conceding that everything is opinion.
The trend in global average temperatures over the past thirty odd years will not be linear. But Tamino’s work has shown that any deviations from linearity in the underlying trend over this period are too small for us ton know their form or direction. They do not rise above the noise.
The methods that he has used are sound. I am a statistician and my judgment here agrees with his.
I’d be skeptical of the Sheldon “method” since it’s demonstrably wrong with this data. But the WUWT blog seems to like it and it reminded me of the GISTEMP results.
1988-1997: +0.03 C/decade
1998-2007: +0.17 C/decade
2008-2017: +0.43 C/decade
I assumed ARMA(1,1) and get an uncertainty in the trend difference is +/-0.38 C/decade from quadrature.
Anyway, using Sheldon’s non-ARMA(1,1) noise model the uncertainty is much less so the conclusion is that warming accelerated through the last decade. The WUWT blog seemed pro-Sheldon technique instead of standard stats, so where did I miss the post reporting accelerated global warming through the last decade?
you sound like a reasonable person. I don’t think that we can be 100% certain of why some things happen, and that is why opinion will always be involved. If there is a slowdown in Gistemp, was it because global warming slowed down, or was it because of an ocean cycle?
The explanation you choose will depend on your assumptions about global warming.
I recognise that Tamino, and probably you as well, have statistics expertise that I don’t have. I am usually willing to listen to other peoples views, when I feel that the person is speaking in good faith (as you said). Having been called a denier for at least the last 9 years, even though I believe most of the same things that warmists do, I feel that being called a denier is an act of bad faith.
I wonder if 2018 will be the year when warmists realise that they have shot themselves in both feet, by calling everybody who disagrees with them on ANY point, a denier.
Tamino’s opinion, and your opinion, may well be correct. But I am not going to give up on the research that I am doing, until I see evidence that convinces me.
This baffoon really does take the biscuit.
♣ Even though he informs us he is fully-signed-up to AGW-inexorably-warming-the-planet and that he is only saying that it is the temperature-wobbles-creating-slowdowns, he then manages to set out this question for us: “If there is a slowdown in Gistemp, was it because global warming slowed down, or was it because of an ocean cycle?” (Beyond the obvious revealing-slip-up, he is probably trying to question the role of statistical values in a model of a physical system and in so-doing forgetting the model in use is dependent entirely on statistical methods.)
♣ He realises he’s been called a denier for almost a decade now. Yet he does not pause (or slowdown) to consider why that might be so. (Hey, you know what? That sounds like somebody in denial, the sort of person you would brand ‘a denier.’)
♣ Even though the baffoon admits both Tamino & Lloyd Flack “have statistics expertise that I don’t have,” he can go no further than see this as ” opinion [that] may well be correct. But I am not going to give up on the research that I am doing, until I see evidence that convinces me.”
♣ And he fantasises that 2018 may be the year “when warmists realise that they have shot themselves in both feet,” apparently by branding all deniers as ‘deniers’. Is he fantasising about being vindicated in 2018? Will 2018 be the year when (à la Month Python) the Black Knight’s defence of his grand bridge (actually a simple plank over a small muddy ditch) will prove to be at last successful? Without arms and legs?
We are good at seeing a pattern on a data set that we are looking at. But we are also good at imposing a pattern when one is not there. The combination of autocorrelation, factors causing short term variation from a trend and a desire to see a particular .pattern make it very easy to see patterns in global temperature data which are not there.
If you want to know whether the trend in temperatures has changed you need a measure of the deviation from linearity of the trend. Then you need to find the probability distribution of your estimate of this measure. After that you need need to calculate the value of this measure on your data set and the probability of getting a value of this measure at least as extreme as the observed one.
You need to describe this deviation in as few parameters as possible if your test is going to be reasonably powerful Tamino has done such tests using a couple of such measures. One was to have a segmented regression. with the measure being the change in slope at a change point. Another was to put a quadratic term into the regression.. Both of these will be simplifications of whatever actually happens but should capture most or at least much of the deviation from linearity in a trend.
So what is your measure of deviation from linearity? Think about it a bit before replying.. On your measure how much of a deviation from linearity matters?. This is relevant to the power of a test. If you want to be able to detect a small deviation then you will need a lot of data and there may not be enough data available to.detect what you want to. If you only want to be able to detect a large deviation then you will need less data.