William M. Briggs

Statistician to the Stars!


Here is the link to the symposium which I mentioned a few weeks back. It is being sponsored by the Ram?n Areces Foundation and the Royal Academy of Sciences of Spain, and will be held in Madrid on the 2nd and 3rd of April. Part of the introduction says:

The Royal Academy of Sciences of Spain and the Ram?n Areces Foundation wish to contribute to the creation of an informed public opinion on global change in the country. To this end, they are organising a two-day symposium aimed at scientists from different fields, decision makers and general public. Existing facts and analysis tools will be discussed, and the robustness and uncertainties of predictions made on the basis of the former, critically assessed. The meeting will provide a scientific view of existing knowledge on climate change and its expected consequences. Existing physical, chemical and mathematical tools will be discussed and climate effects will be analysed together with other concurrent changes, which tend to be overlooked in the climate change scenarios.

Presentations by the different contributors will emphasise existing scientific evidence as well as the strengths and weaknesses of predictions made on the basis of available data and modelling tools. Contributors are encouraged to express their opinions on the most relevant problems concerning the topics they will present, including scientific issues, main threats and possible mitigation or adaptation strategies.

The program is now online. My talk is entitled “Robustness and uncertainties of climate change predictions”. The deadline for me to turn it in is today. I am still working on it and not at all satisfied that I have done a good job with my topic. I am simultaneously writing a paper and the talk, and I will post both of them here, not un-coincidentally, on 1 April.

The gist of my talk I have summarized:

Global warming is not important by itself: it becomes significant only when its effects are consequential to humans. The distinction between questions like “Will it warm?” and “What will happen if it warms” is under-appreciated or conflated. For example, when asking how likely are the results of a study of global warming’s effects, we are apt to confuse the likelihood of global warming as a phenomenon with what might happening because of global warming. When of course the two kinds of questions and likelihoods are entirely separate.

Because of the frequency of confusion, I want to follow the path to the conclusion of one particular study whose results state A = “There will be More kidney and liver disease, ambulance trips, etc. because of global warming.” I start from first principles, and untangle and carefully focus on the chain of causation leading up this central claims, and quantify the uncertainty of the steps along the way.

In short, I will estimate the probability that AGW is real, the probability that some claim of global warming’s effects is true given global warming is true, and the unconditional probability that the effect is true. That’s not too much to tackle, is it?

Thank God there will be simultaneous translation of the conference, because my Spanish is getting worse and worse the more I think about it. If I was going to play soccer, then I’d be on more familiar ground. I do know how to ask that a ball be passed to me because I am alone an unguarded, and how to offer constructive criticism to a fellow teammate for not recognizing this fact and for taking a ridiculous shot at goal himself. But I am not sure how this language would apply to global warming.


  1. Matt:
    It may be late to generate another example, but it strikes me that the a priori probablitlities that people would attach to liver and kidney disease are likely too low and that the import of your analysis will be diluted. If you headline it with a more salient disease like diabetes then you may get additional traction. Obviously you cannot stretch the coverage of the study, but as many AGW proponents have discovered, first impressions may be lasting impressions.

    Good luck with the simultaneous translation – I find them very difficult. My one lesson from doing them in Latin America is to avoid making jokes – many do not translate and one’s sense of humor may not travel far from home!

    Along with everything else, do you still play? My spring season starts in 10 days.

  2. Alan D. McIntire

    March 16, 2008 at 4:12 pm

    I became aware of the “record” problem when I was a freshman at a relatively new high school, only 4 years old when I started attending. Naturally, there were a lot of school records being set each year in various sports. The first year, there’s automatically going to be a new school record in every event measured. Assuming a relatively constant number and relative ability of students each year, the second year about half of the old records will be broken, the third year about 1/3 of the records will be broken, and now that my old high school is about 40 years old, only about 2 1/2 % of school records will be broken in a year. In general, the precentage of records being broken is proportional to the logarithm of the time period.

    If temperature records in the US go back to 1880, roughly 1 in 128 should be broken each year. For a given date, roughly 1 in (128 *365) or 1 in 46,720 should be broken for each date of the year, assuming no trends. With a warming trend, new highs should be somewhat more frequent than that, with a cooling trend, less frequent than that.

    Here I’m guessing, but I suppose the number of new records with a constantly increasing temperature trend should be rougly proportional to
    time / (1 – (T/SD))
    where SD is the standard deviation in temperature
    from year to year, say about 0.1 degrees, and
    T is the linear trend fit, say 0.01 degrees per year.

    In my example, if the world was warming up at 1C per century, and the standard deviation in temperature is 0.1 C per year, you’d expect 10/9
    as many records as the logarithmic prediction.

    With constantly decreasing temperatures, again
    given the above figures, you’d get only 9/10 as many new highs as a strictly logarithmic projection-

    Since the real climate shows increasing from 1880 to 1940, decreasing from 1940 to about 1970, and increasing again from about 1970 to 2000, fudge factor adustments for new records will be slightly more complex than that.
    A. McIntire

  3. I’m confused. Are you saying that agw is real or not or simply being ironic?

  4. Try “Hey, Guyo!”

    Usually works.

    I asked a friend of mine for his opinion on the role of uncertainty a couple of weeks ago. He was at CERN when he got my e-mail. Probably forgotten by now. I’ll have to re-write him. Since his current focus is on strings, the role that uncertainty plays in modeling is something that he has to take into consideration on a daily basis. Something I wish more modelers would do. The most irritating feature of the current debate is the certainty expressed, and criticism supressed.

    It just doesn’t make sense to me.

  5. Speaking as a forester and naturalist, I hold that warmer is better. That is, the effects of GW, should it happen, are likely to be beneficial in the main: longer growing seasons, more biological productivity, more natural wealth creation, less reliance on fuels for heating, more biodiversity, etc.

    At one time (the Eocene) there were boreal tropical forests and many more species than exist today. Indeed, for 99% of the past 300 million years, the Earth has been warmer overall than today. Warmer is the normative condition, if one takes enough paleo-history into account.

    It is in nature of Man that change is held to be bad, precisely because the future is uncertain. Uncertainty itself is held to be a negative. But the mere fact that the future is uncertain does not imply (much less guarantee) that it will be bad.

    Many have expended enormous effort in predicting the terrible consequences of climate change. Few have spent much time making optimistic predictions. However, the future is what we make of it, whatever it might be.

  6. Briggs

    March 17, 2008 at 5:06 am


    Here is my stock statement on AGW

    It is trivially true that man—and every other organism—influences his environment and therefore his climate. It is only a question of how much and to what extent, if any, AGW is harmful or beneficial, and to what extent its harmful effects can be mitigated, or its benefits exploited. We are not interested in trifles: AGW means discernible, large-scale important effects on climate.

    I am trying to understand, or offer a framework for, quantifying the uncertainty of the entire process: how uncertain are our observations, explanations of observations, predictions of new ones, and the effects supposedly caused by AGW.


  7. Ahem, allow me to drop some names. As an undergrad at Berkeley I took Introduction to Statistics from Henry Scheff?. David B. Duncan was a friend of my parents. I was fortunate enough to receive personal lectures from both great statisticians (now deceased), partially dumbed down to my level.

    Their shared interest was compound uncertainty (although they did not entirely agree on the proper stat treatment of compound uncertainty problems). Compound uncertainty may be roughly defined as the increased chance of making at least one mistake when drawing more than one direct inference, for instance when many distributions (or statistical tests) are considered simultaneously.

    Scheff? and others (like John Tukey) considered Duncan’s treatment too “liberal” because it fails to suppress enough Type I errors (false positives), meaning it fails to reject the null hypothesis when the null hypothesis is true. Duncan’s point of view might be crudely summarized as the null hypothesis is never really true, the accused is never really innocent, two phenomena are never really the same, you really are pregnant, despite the statistical significance of the test.

    Both Scheff? and Duncan eventually became Bayesians. Bayesians conjecture that belief systems exist prior to collecting the data, and that those a priori beliefs influence the posterior probabilities that result from data analysis. Another way to say this is that Bayesians reduce the uncertainty through their a priori confidence in their (mystical?) predictive powers.

    I think we are all Bayesians to some degree. We “know” what’s going to happen, or feel like we do. You need some Bayesian confidence just to get out of bed each day and give it another try.

    Alan Watts, on the other hand, (who was not a statistician), preached the wisdom of insecurity. He held that we don’t know shinola about the future, and that’s what makes existence a wonderful adventure.

    I hope all that helps, although I can see how it might not. Here’s another way to look at it: uncertainty aggregates.

  8. Briggs

    March 18, 2008 at 5:49 am


    It’s my belief that everybody’s a Bayesian, but not all admit it. Yet.

  9. I’m ignorant about Bayesianism…and I dislike the segues to philosophy. What I like about Bayesianism is that they want to do bets. However, if I’m at the roulette wheel, my prior is that the numbers will come up the way a frequentist would beleive.

Leave a Reply

Your email address will not be published.


© 2017 William M. Briggs

Theme by Anders NorenUp ↑