Econometric Applications in Climatology

Ross McKitrick is hosting a workshop entitled “Econometric Applications in Climatology” – see website here and here. A detailed program is online here.

Ross has attracted an enviable representation from the econometric community. Invitations were widely extended to the climate community without the response that Ross had hoped for, though there will be some prominent attendees, including Carl Wunsch who will be giving a keynote address.

I’m giving a presentation on Friday on proxy inconsistency at a session chaired by Hu McCulloch. I am consistently amazed at how long it takes me to prepare a new presentation and this has been no exception.


  1. Lance Wallace
    Posted Jun 3, 2013 at 10:25 PM | Permalink

    Too bad the climatologists won’t be there–they might have learned something.

  2. James Smyth
    Posted Jun 3, 2013 at 11:42 PM | Permalink

    I am consistently amazed at how long it takes me to prepare a new presentation

    There is some kind of of converse to “A foolish consistency is the hobgoblin of little minds”.

  3. Geoff
    Posted Jun 4, 2013 at 12:14 AM | Permalink

    Very impressive line up. I see that Dr. Peter Webster (President of the Atmospheric Sciences section of the AGU) is on the organizing committee. Wish I could manage to attend. Good luck on your presentation, Steve. I expect it will be a very valuable conference.

    • Lady in Red
      Posted Jun 6, 2013 at 12:45 PM | Permalink

      Peter Webster is a mentor and friend of Judith Curry. Close support systems are important, I suspect. ….Lady in Red

  4. Posted Jun 4, 2013 at 12:25 AM | Permalink

    Ah, so that’s what you’ve been up to. Good luck with it.

  5. GrantB
    Posted Jun 4, 2013 at 3:45 AM | Permalink

    Since the science is settled it’s hardly surprising the climate community will be under represented. They are the 97% after all.

    • Posted Jun 4, 2013 at 7:17 AM | Permalink

      And no mere econometric application can affect a fundamental constant of nature like that.

      • Posted Jun 4, 2013 at 7:36 AM | Permalink

        The phrase “the usual 97%” in the recent infamous paper had me genuinely LOL’ing at the miraculous depth of “wrongness” it achieves.

        What symbol would represent it? Perhaps ≠ ?

        Back at the workshop, are econometric methods accepted as having legitimate application in what you might call classic climate science?

  6. Posted Jun 4, 2013 at 7:19 AM | Permalink

    I’d love to be at Does Laboratory-Scale Physics Obstruct the Development of a Theory for Climate? by Christopher Essex. All the very best with this Ross and Steve.

    • Steven Mosher
      Posted Jun 4, 2013 at 9:50 AM | Permalink

      Hmm, having read the paper I’d beg to differ

      • Posted Jun 4, 2013 at 11:39 AM | Permalink

        Reading papers again. You really have to kick that habit you know 🙂

        • Steven Mosher
          Posted Jun 5, 2013 at 12:29 PM | Permalink

          it was nice that they posted them. I was hoping steve’s was there.

      • Geoff Sherrington
        Posted Jun 6, 2013 at 2:49 AM | Permalink

        Agreed, Steven, after reading the short version.

    • Posted Jun 4, 2013 at 1:05 PM | Permalink

      I’d like to hear about the talk on stationarity also.

      • Posted Jun 4, 2013 at 1:18 PM | Permalink

        There will be a number of papers taking different positions on various aspects of the stationarity issue. I expect it will be the first time a large group of capable time series experts will be brought together to debate whether climatic data should be considered nonstationary and how trends ought to be modeled. Given the implications it seems to me such a debate is long overdue.

        • Geoff
          Posted Jun 4, 2013 at 7:32 PM | Permalink

          This new paper may be of interest

        • Geoff Sherrington
          Posted Jun 7, 2013 at 7:59 PM | Permalink

          A Conference like yours has long been needed and I hope it filled your expectations. Thank you. Will there be a lifting of the embargos on papers at a known date?
          Thank you also for the pointer to Hlinka at. al.

          My mental problems start right from the beginning. In terms of purity rather than semi-quantitative practicality, I have asked a few people the basic question that follows, but no answer yet seems to satisfy. Perhaps readers could keep it in mind as they read the Conference papers.

          Some autocorrelation properties of Tmax differ from those of Tmin at a location, as I have shown by a short, rough, unfinished essay using daily lagged temperature data for Melbourne.

          It is common practice to create a Tmean from (Tmax+Tmin)/2. In routine practice, groups such as the Australian BOM create a CLIMAT report each month and send it to global bodies like Hadley and NOAA. Different ones of these use Tmax and Tmin to create their Tmeans.

          The fundamental question is whether it is valid to use the Tmean created by combination of Tmax and Tmin, which have different properties, for further statistical work, such as further autocorrelation math.

          Additional information. A private email from BOM notes “CLIMAT data feeds are not based on (min+max)/2, they are based on whatever the country supplies. I did a survey of this as part of the introduction to my 2004 paper and found that, roughly speaking, 40% of the world’s area uses (min+max/2) (mostly English-speaking countries), 40% uses the mean of evenly-spaced observations (the largest countries that do this being China and the former USSR), and 20% use more complex formulae which attempt to replicate a ‘true’ mean through weighted averages of temperatures at particular hours (mostly continental Europe and Latin America). “

          I was reminded of a famous W. M. Briggs quote “Now I’m going to tell you the great truth of time series analysis. Ready? Unless the data is measured with error, you never, ever, for no reason, under no threat, SMOOTH the series! And if for some bizarre reason you do smooth it, you absolutely on pain of death do NOT use the smoothed series as input for other analyses!”

        • ferd berple
          Posted Jun 17, 2013 at 8:10 AM | Permalink

          Re: Ross McKitrick (Jun 4 13:18),
          whether climatic data should be considered nonstationary
          It is irresponsible if this issue hasn’t been addressed by climate science. Climate is the statistics of weather. If weather is not stationary, then you cannot apply linear regression (trends) to the data directly, you must first difference the data. Thus, all the time series graphs of temperate we see with trend lines on them would be statistical nonsense. Which includes just about every graph on AGW ever published.

          Given that it has long been known that tomorrow’s weather is more likely to be like today’s weather than otherwise, and that one can more often replace tomorrow’s weather forecast with today’s weather and improve accuracy, this would seem to strongly argue that weather is not stationary. Imagine you had a coin that was more likely to throw heads if the last toss was heads. If the percentage of heads is increasing over time, is it due to AGW, or is it simply due to the nature of the coin?

          Then there is the issue of trying to average a chaotic system over time. The law of large numbers is not on your side with chaotic systems. The “noise” doesn’t average out to zero over time, because it isn’t noise it is chaos. What looks like an average is simply a local attractor, and over time you are likely to wander towards a different attractor, throwing your carefully built average out the window and rendering it nonsensical. Unlike a coin toss, your chaotic system will not converge.

          As you increase the length of the series, the average will diverge. Over 60 years temperature has been increasing, over 6000 decreasing, over 60,000 increasing, over 60 million decreasing. All these series yield different average temperatures. Which one is the true average temperature of the earth?

  7. Jeff Norman
    Posted Jun 4, 2013 at 1:48 PM | Permalink

    “”Invitations were widely extended to the climate community without the response that Ross had hoped…””

    As pretty as Guelph may be in June, it is in no way comparable to Bonn, Mauritius, Doha, Chicago, Rio de Janeiro, Cancun, Copenhagen, Bali, Nairobi, Montreal, Moscow, New Dehli, etc.

  8. Posted Jun 4, 2013 at 7:26 PM | Permalink

    I ewnquired about whether they had taken into account Dr Bouldin’s critique of dendroclmatology….atill waiting to see it in the borehole

  9. David L. Hagen
    Posted Jun 4, 2013 at 7:56 PM | Permalink

    Two papers on Granger / cointegration

    Polynomial Cointegration Tests of Anthropogenic Impact on Global Warming, Yaniv Reingewertz

    Long-memory and the Sea Level-Temperature Relationship: a Fractional Cointegration Approach
    <a href= Ventosa-Santaularia

    For background, see: Beenstock, Reingewertz, and Paldor “Polynomial cointegration tests of anthropogenic impact on global warming”
    5 citations

    D. Stockwell discusses cointegration including Beenstock et al. and
    sea level

    And while the correlation is high when sea level is added into the linear model, the sea level almost blocks out all the other variables:

    May the sparks fly!

    • David L. Hagen
      Posted Jun 4, 2013 at 8:10 PM | Permalink

      mae culpa: see <a href= Ventosa-Santaularia‘s papers

  10. Posted Jun 5, 2013 at 10:01 AM | Permalink

    “There is no more common error than to assume that,
 because prolonged and accurate mathematical calculations have been made,
the application of the result to some fact of nature is absolutely certain.”
- A.N. Whitehead, English mathematician and philosopher 1861-1947

  11. MikeN
    Posted Jun 5, 2013 at 10:24 AM | Permalink

    Will VS be there?

    • Posted Jun 5, 2013 at 10:50 AM | Permalink


      • Posted Jun 5, 2013 at 5:24 PM | Permalink

        Best wishes to you Ross for a really successful — first of its kind — conference. And best wishes to Steve on his presentation at the conference. The topic chosen seems to be very interesting and right up his street. What better place to hold the meeting than at UofG — my o;d stomping ground.

      • kim
        Posted Jun 5, 2013 at 11:29 PM | Permalink

        Bart V?

      • cartoonsbyjosh
        Posted Jun 10, 2013 at 7:57 AM | Permalink

        Wish I could be there to meet everyone and take notes. I will be with you in spirit… and in cartoon 😉

  12. Lady in Red
    Posted Jun 6, 2013 at 12:54 PM | Permalink

    I am *so* pleased Carl Wunsch is becoming involved. I’ve been water torturing him for over a year with snippits from Climate Audit and Climate Etc., begging him to get involved, help the science regain honesty.

    (Wunsch was lecturing at Oxford when the Peter Gleick mess exploded — and, in the middle of it all, Gleick was invited to speak there. I never heard a word about that from Wunsch, but….)

    I am very happy. There are other impressive scientists, grey beards especially, who can be cajoled back to the light, advocacy of honest science. ….Lady in Red

    • Steve McIntyre
      Posted Jun 6, 2013 at 1:40 PM | Permalink

      Wunsch thus far has been a no-show. He says that he suddenly got food poisoning.

      • Lady in Red
        Posted Jun 6, 2013 at 1:44 PM | Permalink

        Damn. I sent him a thank you.

      • kim
        Posted Jun 7, 2013 at 3:10 AM | Permalink

        er, Durkin. I wonder how Carl feels about ‘An Inconvenient Truth’?

      • Posted Jun 8, 2013 at 3:08 PM | Permalink

        Carl made a sincere effort to make it but he became ill just before the conference and did not improve during the week. We were very disappointed that he could not attend, as he was at missing the meeting.

        The workshop went very well. The quality of presentations was extremely high, and the debates and discussions were fascinating.

        • Posted Jun 9, 2013 at 9:47 AM | Permalink

          Respect for truth making a comeback on both sides of the Atlantic this weekend? Congratulations Ross.

        • Ivan Jankovic
          Posted Jun 15, 2013 at 11:16 AM | Permalink

          Did you invite James Hansen to participate?

    • Posted Jun 7, 2013 at 12:50 AM | Permalink

      Lady in Red –

      Carl Wunsch is one scientist I would like to have a beer with sometime. Though he seems to be sold on the CO2 as the devil and that global warming is a serious catastrophe we need to do lots to avoid, on one point I love his position. And that is the oceanic conveyor hypothesis.

      I think that is one of the looniest ideas ever to gain wide support. I have written elsewhere how the idea is simply over-the-top stupid, on several fronts. I think the people pushing it (and they seem to be everywhere) have no idea about how currents are driven, how the Gulf Stream works, and how suction and convection work.

      So when he writes this I feel vindicated:

      I am on record in a number of places as complaining about the over-dramatization and unwarranted extrapolation of scientific facts. Thus the notion that the Gulf Stream would or could “shut off” or that with global warming Britain would go into a “new ice age” are either scientifically impossible or so unlikely as to threaten our credibility as a scientific discipline if we proclaim their reality.

      Very strong words. I think I might post on my small blog about why this all is impossible. But give me a while to put it up.

      In general, good scientists (unlike lawyers) are meant to keep in mind at all times that conceivably they are wrong. There is a very wide spectrum of scientific knowledge ranging from the almost certain,. . . to inferences that seem very plausible but for which one can more readily imagine ways in which they might prove incorrect . . . to fiercely disputed ideas (e.g., that variations in the North Atlantic circulation directly control the climate of the northern hemisphere).

      That last – I do truly HOPE that someone out there is fiercely disputing it. And I am glad that Dr. Wunsch is one of them.

      Please keep after him. He actually seems like a scientist who recognizes at least SOME bad science when he sees it. Yes, as in “honest science.” Keep cajoling him. He would be a major coup to have him switch sides.

      • Harold
        Posted Jun 13, 2013 at 5:59 PM | Permalink

        But Bill Nye said the thermohaline thingy was going to do something.

      • Posted Jun 14, 2013 at 8:35 PM | Permalink

        [Wunsch] actually seems like a scientist who recognizes at least SOME bad science when he sees it

        I hope he makes a speedy recovery from that which prevented him from participating in Ross’s conference. And when he does, I hope he will take a second look at a (non-scientific, to my eyes at least) “statement” he has endorsed (along with a number of less surprising names: e.g. Mann, Gleick, Hansen, Santer, Weaver, Karoly, Ehrlich and Suzuki) See:

        Crisis of the week: the biosphere … new “Statement” percolated, circulated and endorsed

  13. Lady in Red
    Posted Jun 6, 2013 at 1:46 PM | Permalink

    Perhaps “other forces” have been at work. I think his conscience is itching.

    Pls keep, gently, after him, ….in the name of honest science.

    • Steve McIntyre
      Posted Jun 6, 2013 at 2:05 PM | Permalink

      it’s not a “skeptic” workshop. Most of the people are econometricians.

      • Lady in Red
        Posted Jun 6, 2013 at 2:57 PM | Permalink

        I know. ….smile.

      • Posted Jun 6, 2013 at 7:22 PM | Permalink

        Econometricians who like data, are probabilistically inclined to be deemed skeptics.

  14. johanna
    Posted Jun 7, 2013 at 12:03 AM | Permalink

    Look forward to reading your report from the conference Steve (if you have time) and also to the conference report from the organisers. Any chance of a guest post from Ross McKitrick summarising the highlights?

    I’d love to be there, but since I am not on the gravy train, airfares and accommodation from Australia to Canada for a conference are beyond my means.

    Econometrics is not a bad “proxy” for climate science, in the sense that it involves modelling complex dynamic systems. Definitely a worthwhile exercise.

    • Guy Leech
      Posted Jun 7, 2013 at 9:37 AM | Permalink

      Econometrics is also similar to climate science in that it is often not very successful at forecasting the future.

  15. Ivan Jankovic
    Posted Jun 13, 2013 at 11:26 AM | Permalink

    It would be interesting why the econometricians think about this:

    • Ivan Jankovic
      Posted Jun 13, 2013 at 11:48 AM | Permalink

      ‘what’ not ‘why’ 🙂

      • Posted Jun 18, 2013 at 9:48 AM | Permalink

        Ivan Jankovic.
        Murry Salby is no longer at Macquarie Univeristy. Do you know where he is located.

%d bloggers like this: