A Random Thought on the Scientific Method

In response to this post, climate model skeptic Rick Stryker writes (in ALL CAPS no less):

JUST BECAUSE A MODEL DESCRIBES THE EXISTING DATA DOESN’T MEAN THAT IT WILL DESCRIBE DATA THAT HAS NOT BEEN OBSERVED

So far we’re in agreement; in fact I’m going to repeat this point to my econometrics class. He then continues:

You see, in science, you don’t prove the theory by showing that it describes the existing observations. You prove the theory by showing that it predicts data that haven’t been observed.

Well, gee, if this is the standard for proving or disproving hypotheses, either generally, or in econometrics, we’re not going to get very far. In this view, I won’t see our sun go nova, so might as well call it a day — science can’t proceed until we get the data! But this is the sort of nihilistic worldview that pervades the global climate change deniers.

For a more succinct critique, see below:

toles_kyoto

StumbleUponLinkedInReddit

87 thoughts on “A Random Thought on the Scientific Method

  1. CoRev

    Menzie, please show us the denier in the subject thread which supports your denier</B. analysis. This comment: " But this is the sort of nihilistic worldview that pervades the global climate change deniers." is just your personal projection. Blind belief does not make a subject fact, and that's all you are demonstrating while disparaging those who better understand or at least a different understanding of the issues.

    The other thread proves that there are many different issues for the science, and to assume that it is settled and policies for solving it are issue free is arrogant and ignorant. You had an opportunity to discuss the science in the previous thread and eschewed it.

  2. baffling

    corev,
    your problem is you think that science is a debate, and if you talk long and loud enough the other side will quit and your view will rule. science is not a debate. i understand you don’t believe in climate change. i understand you don’t believe in evolution. heck, i even understand your frustration with the law of gravity. but you know what, even when i quit trying to convince you gravity exists you have not won. because when you step out onto that sheet of ice you will still fall flat onto your back because gravity still exists in spite of your debate against it!

    1. CoRev

      Baffled, I have no problem, I just have a better understanding of the nuances of climate science than the average blogger.
      Joseph, perhaps you missed my typing error in not closing out the bold. Did you actually have a point?
      Jon Hall, yes they do try to correct measurement errors, data drop outs, failed coverage in area and time, and even Urban heat Island effects (site location biases). In the end we see error corrections greater than the signal. That adds to the distrust, because overtime the data shifts change the original data, because nearly all the corrections add to lowering older temps. Old lower temps compared with warmer new temps leads to disputed warming trend values.

      Even the new IPCC report will report some lowering in previous report climate sensitivity findings, and that is due primarily to the acceptance of the hiatus.

      1. kharris

        “I just have a better understanding of the nuances of climate science than the average blogger.”

        OK, in the standard list of logical fallacies, “argument from authority” has a respected position. We apparently need a special spot for CoRev – argument from assumed authority. CoRev claims he knows best, and then relies on his own views as authoritative. Long has it been thus.

  3. Joseph

    Corev, could you speak a little louder? Boldface won’t do it. All caps is better. Best would be boldface, italic, all caps.

  4. John Hall

    My understanding is that some of the issue is that there isn’t as much new accurate data and the old data is sort of noisily measured. It seems like the perfect application of Bayesian techniques. I’m not that familiar with the literature on climate warming to know if they explicitly incorporate measurement error in their estimates.

  5. Steven Kopits

    “Well, gee, if this is the standard for proving or disproving hypotheses, either generally, or in econometrics, we’re not going to get very far.”

    In fact, economics is not getting very far. There seems to be no general agreement about whether the fiscal stimulus worked, about fiscal multipliers, and even whether raising the minimum wage is likely to lead to greater unemployment. It is a sorry state of affairs, which I think economists’ wide scale blogging has made more acutely evident.

    I agree with Rick that the value of any model is in its ability to predict future events, although the unfolding of events as predicted is not proof a model was right–only that it was useful.

    Notwithstanding, we are compelled to take action on such information as is available. The greater and more certain the potential cost, and the more uncertain the benefit, the greater resistance to change and the greater desire to see incoming data conform to predictions, or not, as the case may be. With regards to climate, the data has been coming in well below most projections, thus skeptics feel quite justified in wanting more data and deferring action.
    http://wattsupwiththat.com/2014/02/20/a-must-read-why-secretary-of-state-john-kerry-is-flat-wrong-on-climate-change/#more-103569

    We often see ex-post forecasts. “See, global warming is leading to this bitterly cold winter!” But this never seems to be forecast in advance. Instead, back in 2000, Dr David Viner, a senior research scientist at the climatic research unit (CRU) of the University of East Anglia said that, within a few years winter snowfall would become “a very rare and exciting event”. “Children just aren’t going to know what snow is,” he said. Perhaps British children should come to Princeton. We can show them what snow is! But I don’t see any disavowal of Viner. Did any alarmists say, “Gee, David is full of it. We’re still going to get plenty of snow.” I don’t recall that.

    http://www.independent.co.uk/environment/snowfalls-are-now-just-a-thing-of-the-past-724017.html

    1. anon2

      As I said in the prior thread: I think the sticking point in all this arguing over global warming is whether people can admit that human activities do indeed impact the larger world in significant ways. Once you realize that, then the next step, unfortunately, is how to regulate human actions to keep the planet livable, a step “conservatives” do not want to take.

      A few anecdotes does not prove the science wrong.

      But what never ceases to amaze me is that the climate warmers seek to find a path to a sustainable future, and the climate deniers think that’s hogwash. It always reminds me of the story of the grasshopper and the ants. You know who wins.

      1. aaron

        “I think the sticking point in all this arguing over global warming is whether people can admit that human activities do indeed impact the larger world in significant ways. ”

        You also need to determine the impacts and whether they can be adapted to or not. An unintended consequence in not necessarily a bad thing.

      2. Steven Kopits

        Whoa, Anon2. No one said humans don’t affect the planet. Of course we do.

        Skeptics do not at all reject the notion that CO2 has increased, and that at least half of the increase is anthropomorphic. Nor do skeptics debate whether CO2 increases temperatures. It does, or at least it should.

        The issue is whether the increase is material and worth doing something about, when the proposed remedies are materially damaging to human economy or society. There skeptics would argue that the cost benefit of doing away with fossil fuels is probably not worth it. That’s the active point of contention.

        1. baffling

          kopnits,
          “Skeptics do not at all reject the notion that CO2 has increased, and that at least half of the increase is anthropomorphic. Nor do skeptics debate whether CO2 increases temperatures. It does, or at least it should.”

          this is where you are wrong. when the debate first began many years ago, skeptics insisted humans could not influence the climate of a massive object such as earth. while they have slowly, and begrudgingly, evolved their stance in the presence of evidence, the fact remains you have a sector who really does not believe in science. in fact they believe science is a debatable subject-if i argue long and loud enough my theory of science will win. and this has been detrimental to the development of climate science because it has had an negative impact on the ability to conduct impartial research, due to politics and economics. while i tend to disagree with you on topics, but you appear to be the minority in a group of folks who probably would change their mind on the topic when evidence exists. but that is not the typical perspective of the “denier” crowd scientists must contend with.

          1. CoRev

            Baffled, another strawman argument with no support except your own opinion. ?Denier?, ” science is a debatable subject?” since when had scientific debate been outlawed?, “does not believe in science”, Huh? A whole set of misconceptions based upon — opinion.

            KHarris, where’s your arrogance when we see this kind of foolishness?

        2. Nick G

          Skeptics do not at all reject the notion that CO2 has increased, and that at least half of the increase is anthropomorphic. Nor do skeptics debate whether CO2 increases temperatures.

          If only that were true. In fact, the history of resistance to the ideas of Climate Change are full of examples of exactly these ideas.

          the proposed remedies are materially damaging to human economy or society.

          That is misinformation, which comes from those who are resisting change. In fact, the proposed remedies (efficiency, electrification of transportation, etc) would have both short term and longterm economic benefits.

          For instance, car makes have fought desperately against increases in the CAFE regulation, when such improvements would save consumers enormous amounts of money over the lifetimes of their vehicles. You have presented evidence to support this, by presenting yourself as an example of a consumer, who discounts the future so excessively that they demand a 33% ROI from efficiency investments.

    2. floyd

      Snow Mistake?
      Long ago I read about a scientist (studying cores from, I believe, the English Channel) who had come to the conclusion that there had been period(s) when excessive
      snowfall had led to periods of non-total snow melt summers. The “snow” scientist could be mistaken in the opposite direction. Heavy snow falls in relatively warm winter periods. Thus global warming could result in re-glaciation.
      Equilibrian Earth.
      Floyd

    3. Nick G

      The greater and more certain the potential cost, and the more uncertain the benefit, the greater resistance to change

      That’s not the problem. The problem is that the cost falls on a narrow part of society (investors and workers in industries that emit CO2), and the benefits are much more widely spread. If you’ve invested your life’s work (or your money) in oil or coal (or internal combustion engines, etc.) then you’re going to fight the transition away from fossil fuels. That’s easy to understand, and one can have compassion for those who are caught in such a bind. Still, the larger society can’t avoid the transition. Ideally, the larger society would find a way to soften the blow. But, the blow must come, and the sooner the better.

      The cost of a transition away from fossil fuels is, in fact, much smaller than the cost of staying with them. Even the immediate, obvious costs of fossil fuels are higher than the costs of alternatives. But, those who benefit from fossil fuels wish to misinform us on this topic, purely to protect themselves.

    1. Lyle

      Reading the 1960 Britannica and then comparing the modern Wikipedia definition: http://en.wikipedia.org/wiki/Heresy
      “. Scientist/author Isaac Asimov considered heresy as an abstraction,[40] mentioning religious, political, socioeconomic and scientific heresies. He divided scientific heretics into endoheretics (those from within the scientific community) and exoheretics (those from without). Characteristics were ascribed to both and examples of both kinds were offered. Asimov concluded that science orthodoxy defends itself well against endoheretics (by control of science education, grants and publication as examples), but is nearly powerless against exoheretics. He acknowledged by examples that heresy has repeatedly become orthodoxy.” From the section on non religious heresy. It is clear that there is an orthodox belief in AGW, and to these folks the deniers are heretics,. when one says that the consensus is xyz then one implicitly has said it is the orthodox position. Since many with the orthodox position think the problem could be the end of the world (at least as we know it), they come out with religious fervor, as do the deniers who have their own orthodoxy.

      1. CoRev

        Lyle, Kopits comment her to get a feel for how you have misstated the issue. There is a well accepted belief in AGE.
        Its not just: “there is an orthodox belief in AGW,”. Indeed the 97% studies showed that, and you highlight a fundamental difference between the two beliefs when you say: “think the problem could be the end of the world (at least as we know it),”. A skeptic interprets that as “are you really trying to stop change? And where is there any evidence of the catastrophic capability of warming?

  6. George Phillies

    The prediction of noisy but advancing warming, as witness by the incredibly warm winter (if you are in Alaska) seems to be doing quite nicely. The most visible failings are that the warming and, e.g., loss of Arctic ice cap in Summer, have been more rapid than anticipated.

    1. aaron

      Yes, but arctic ice has increased recently and we don’t have good historical data. Arctic sea ice age is young, suggesting that it has been low in the past. Given that arctic ice was probably unusually high due to the mini-iceage, the young ice age is especially indicative of oscillation.

  7. Bruce Hall

    If the scientific method mean “the science is settled,” then perhaps it is time to start using another method. Did Newton “settle the science?” Did Einstein “settle the science?” Do mere statistical correlations “settle the science?” And since when did models that do not predict observable phenomenon “settle the science?”

    The dumbest phrase ever uttered by a scientist: “the science is settled.”

  8. Brian

    I always thought the scientific method included having a falsifiable hypothesis. With the theory of gravity, I can say that if I drop a lead ball and it goes upward and smacks me in the face while standing on the ground with no wind, the theory is false.

    With climate change advocates, just about everything is claimed to be consistent with the theory.

    I simply want to know what physical observations would NOT be consistent with AGW theory. Is there anything at all that would cast doubt on the theory?

    1. libert

      Per your point about the scientific method, recall the original objection: “JUST BECAUSE A MODEL DESCRIBES THE EXISTING DATA DOESN’T MEAN THAT IT WILL DESCRIBE DATA THAT HAS NOT BEEN OBSERVED”

      Just because the theory of gravity describes the existing data (i.e., things have always fallen in the past) does not mean that it will describe data that has not been observed (i.e., whether or not things will fall in the future).

  9. Patrick R. Sullivan

    Related to the previous thread on the minimum wage, but appropriate to any discussion of proper science, is this 2013 paper by Meer and West of Texas A&M;

    http://econweb.tamu.edu/jmeer/Meer_West_Minimum_Wage.pdf

    ‘Granted, there remain numerous state-year increases in the minimum wage that were never fully eroded by inflation or, in a relative sense, by neighboring states later boosting their minimum wages. However, this exercise demonstrates that there is a relatively short duration of time during which a state difference-in-differences estimation can identify the effects of the minimum wage on employment levels. This situation would not be problematic if the minimum wage affected employment in an abrupt, discrete manner. But if the minimum wage predominantly affects job creation, then it may take years to observe a statistically significant difference in total employment.8 Thus, while it is true that any reduction in job growth should be reflected eventually in total employment, the empirical challenges discussed in this section may preclude identifying the net effect of the minimum wage by examining employment levels directly. As a result, even though the employment level is the outcome predominantly considered in the empirical literature on the minimum wage, we focus instead on the job growth rate.’

    That pretty much devastates the Card-Krueger approach to measuring employment effects of the minimum wage.

  10. alex

    Your supernova comment is a straw man. The laws of physics are not only consistent with an enormous accumulated body of experimental results, they also correctly predict the results of (all) new experiments every day. The physics that describe supernovae is the same physics that governs terrestrial collider experiments and it correctly predicts the results of new experiments every day. Climate models are different: As we keep hearing, “weather is not climate”. That’s because the timescale for climate is LONG. So we have had approximately ZERO out of sample climate observations, given that the climate models have their parameters chosen by fitting to historical data up to the present. You, of all people, should be familiar with this issue: Econometric models, like climate models, work fine for ex-post retrodiction but we know that they’re pretty useless for prediction. I can give you half a dozen very reasonable, parsimonious, economically well-grounded models that suggest investment strategies that would have had spectacular information ratios. But that’s only true in-sample. In real-world, out-of-sample forecasting they all have close to zero forecasting power. Markets, like climate, have feedbacks and a variety of timescales and are just too complex to believe in any given model’s forecasting power. No matter how good the in-sample fit is. Same with climate.

    And, finally, let me quote from section 9.2.3 of the IPCC report: “The approach to model evaluation taken in the Chapter reflects the need for climate models to represent the observed behaviour of past climate as a necessary condition to be considered a viable tool for future projections. This does not, however, provide an answer to the much more difficult question of determining how well a model must agree with observations before projections made with it can be deemed reliable. Since the AR4, there are a few examples of emergent constraints where observations are used to constrain multi-model ensemble projections. These examples, which are further discussed in Section 9.8.3, remain part of an area of active and as yet inconclusive research.”

    So there :D.

  11. 2slugbaits

    aaron arctic ice has increased recently

    Where are you getting this nonsense?

    https://nsidc.org/arcticseaicenews/

    Steven Kopits The issue is whether the increase is material and worth doing something about, when the proposed remedies are materially damaging to human economy or society. There skeptics would argue that the cost benefit of doing away with fossil fuels is probably not worth it. That’s the active point of contention.

    Not quite. The issue is whether you give any weight to the welfare of future generations. The proposed remedies to global warming are indeed damaging to current and near term generations. The next couple of generations will pay some cost (perhaps 2% of GDP) and receive very few benefits from reduced CO2 emissions. But ignoring global warming today quite literally dooms future generations to an unimaginably horrible world. So if you don’t want to pay a small price today, then your grandchildren’s grandchildren will pay dearly in the future. The only people who believe the cost/benefit of reducing our dependence on fossil fuels is not worth it are people who quite literally put zero value on the welfare of future generations. A lot of these smokescreens about “unsettled science” and “climate skepticism” are just self-serving rationalizations to avoid having to answer the question about how guessing wrong about global warming will impact future generations. The core problem is that what we do today to fix global warming will not directly benefit us; and what we fail to do today to fix global warming will hurt future generations. So this is really a representative agent problem across generations. And humans are not very good at solving those kinds of problems.

    Also, no one is saying that we need to do away with fossil fuels. We do need to leave 80% of the carbon in the ground. How we allocate the remaining 20% that we can consume is an economic question. Ultimately we will need to develop more solar and wind power, but I don’t think those two sources will be anywhere near adequate. The only real long run solution is to develop 4th generation breeder reactors. Unfortunately they are several decades away from going prime time, so we need to get busy on that project. But that would also mean bucking Koch Brothers coal.

    1. Nick G

      The proposed remedies to global warming are indeed damaging to current and near term generations.

      It’s a mistake to stipulate to such an argument. In fact, oil is very expensive compared to the alternatives. New land-based windpower is cheaper than new US coal plants (because they scrub sulfur, mercury, etc). The UK is finding windpower expensive only because they’re not willing to install land based wind turbines – apparently they’d rather spend a lot more money putting turbines out to sea. Of course, if fossil fuel interests weren’t encouraging such astroturf, the UK would have a more sensible policy.

      solar and wind power, but I don’t think those two sources will be anywhere near adequate.

      Don’t concede this either. Solar can easily provide 35%, and wind can provide 50% of what’s needed. In the long run, of course, solar can provide far more than 100%, although that may not be the cost optimal solution. And, of course, fission reactors will work if needed.

  12. Rick Stryker

    Menzie,

    I guess the all caps didn’t help since you misunderstood the point anyway. For your benefit and others, I’ll write the argument out in great detail so it is very clear.

    The most sophisticated models that global warming alarmists use to make their dire predictions are called atmosphere-ocean general circulation models (AOGCM). These models divided the surface of the earth into small blocks and the atmosphere and oceans into layers. In these models, the atmosphere, oceans, and surface of the earth are coupled into a 3-dimensional climate model. The model’s calculations describe the flow of heat and fluids (eg ocean water and air) as they flow into and out of the blocks and layers. The calculation burden of these models is enormous and so the models are not solved starting from the past and propagating to the present. Instead, the models start from the features of the world today, perturb the environment with greenhouse gases, and then calculate the effects on the temperature, atmosphere, oceans, clouds, ice, and snow in a time step. Then a further perturbation is added and everything in the model is recalculated. The process continues as the model steps through time.

    A very important feature of these models is the inclusion of adjustable parameters which determine the strength of some of the processes in the model. These parameters are tuned so that the model can reproduce the already observed data such as the historical temperature data. As I mentioned in the comment that prompted your post, these tuned parameters, necessary to match the past data, produce the positive feedback that is responsible for the alarming predictions of these models.

    Now to my criticism. You seem to think my point about predicting unobserved data illustrates the “nihilistic worldview that pervades the global climate change deniers.” But the essential features of my point are acknowledged by the authors themselves of the Intergovernmental Panel on Climate Change (IPCC) report. Take a look at chapter 9, Evaluation of Climate Models in the 2013 IPCC report. Allow me to quote from “FAQ 9.1 | Are Climate Models Getting Better, and How Would We Know?” on page 824:

    “An important consideration is that model performance can be evaluated only relative to past observations, taking into account natural internal variability. To have confidence in the future projections of such models, historical climate—and its variability and change—must be well simulated. The scope of model evaluation, in terms of the kind and quantity of observations available, the availability of better coordinated model experiments, and the expanded use of various performance metrics, has provided much more quantitative information about model performance. But this alone may not be sufficient. Whereas weather and seasonal climate predictions can be regularly verified, climate projections spanning a century or more cannot. This is particularly the case as anthropogenic forcing is driving the climate system toward conditions not previously observed in the instrumental record, and it will always be a limitation.”

    As you can see, the authors are fully acknowledging that although the models fit the data in sample, out of sample predictions out to 100 years are not verifiable. Further, they go on to say that “this alone may not be sufficient” or in my language–just because a model predicts in sample doesn’t mean it can predict out of sample. How then can we have confidence in these models? Let’s skip down the page for the answer:

    “So, yes, climate models are getting better, and we can demonstrate this with quantitative performance metrics based on historical observations. Although future climate projections cannot be directly evaluated, climate models are based, to a large extent, on verifiable physical principles and are able to reproduce many important aspects of past response to external forcing. In this way, they provide a scientifically sound preview of the climate response to different scenarios of anthropogenic forcing.”

    In other words, they are saying that since the model can reproduce the past more or less and the since model is based on the known physics and chemistry, we should therefore trust it to make long horizon predictions which they themselves acknowledge can’t be verified and aren’t necessarily implied by the ability to predict in sample.

    Now there is an obvious problem with their argument. Maybe the fact that the models can reproduce the past more or less means that they got it right and the model is sound. But it’s also possible that they got it wrong but that the fudge factor tuning parameters are disguising that fact. How do we know?

    Well, it turns out that we already have some evidence which is contained in this very document. Global warming alarmists are still partying like its 1999, but it’s hard to see how they can keep the merriment up given the elephant in the room: since around 1999 there has been a lull in the temperature increase that the models have not been able to explain. Let’s skip over to page 769 entitled “Box 9.2: Climate Models and the Hiatus in Global Mean
    Surface Warming of the Past 15 Years” in which the authors explain the problems the models have accounting for the 15-year temperature pause. In the subsection entitled “Internal Climate Variability” they make a critical admission, although it’s buried:

    “Owing to sampling limitations, it is uncertain whether an increase in the rate of subsurface–ocean heat uptake occurred during the
    past 15 years (Section 3.2.4). However, it is very likely that the climate system, including the ocean below 700 m depth, has continued to accumulate energy over the period 1998–2010 (Section 3.2.4, Box 3.1).”

    Going back to my description of the models, that statement means the models fail to simulate the transmission of heat from the upper layers of the ocean in the model to the layers in the deep ocean. That’s a problem and it raises the question of what other features might be missing or wrong in the models.

    The global warming modelers want us to trust that they got the physics right since they can reproduce past data. But that’s not really an acceptable answer since the tuning parameter fudge factors can hide mistakes in the models. So, going back to my point, I was saying that to trust these models they must be able to make non-trivial predictions about some data that can be observed but hasn’t been observed, since if the data has already been observed the model has already been tuned to account for it. I don’t mean that the models should predict data 100 years from now, since that is clearly useless, but rather non-trivial data that can be verified through some measurement or experiment.

    Until these models can meet that standard, there is no reason to put much trust in them. And since global warming alarmism depends on the correctness of the models, there is no reason to put much trust in global warming alarmism. We should all relax. There is no emergency.

    1. CoRev

      Rick Stryker, excellent job. Your conclusion is right on ” And since global warming alarmism depends on the correctness of the models, there is no reason to put much trust in global warming alarmism. We should all relax. There is no emergency.” I would add there is no record of any of the extremes upon which this alarmism is based.

      All the climate hype is man made and from the models. Worse, as I said earlier, much of the observational evidence is from the adjustments to the observations. When they exceed the signal, minor changes causing adjustment errors perturb the observation data such that even they can not be fully trusted. What we have seen is the vocal group of climatologists running the older established datasets argue the new satellite and ARGO buoy data into compliance with their own trends if they diverge. So we end up with group think compliance in the observational data. That too is happening with the models. Will the diverging model(s) be funded to the same level when its results are questioned by the vocal few?

      In time I think we will find Climate Science becoming a case study for group think, funding impacts, and human nature in the social sciences and not hard science

      2slugs, just or you, denying the data of a hiatus is an example of this religious fervor. It isn’t CoRev’s hiatus!

      The elephant in the room, the hiatus, has been predicted for years by many skeptical scientists. Skeptics have been able to more accurately predict future trends than have the orthodox/traditional science community. That group has relied on models-based estimates/predictions and mostly ignored the uncertainty surrounding them. Most of the arguments ignore the uncertainty, and as I claimed there is uncertainty in both the science and the observations. What we see in the masses is emotions verging on religious fervor based upon these extreme, uncertain and unsupported model predictions.

      And all of this arguing surrounds a historical change of less than 1 degree centigrade happening over the equivalent three generations of history. All of this arguing is over the latest data points in the latest peak of multiple cyclical peaks and valleys where the peaks are diminishing. The Holocene story: http://jonova.s3.amazonaws.com/graphs/lappi/gisp-last-10000-new.png

    2. baffling

      rick, is your 15 year “hiatus” the trend or statistical noise, ie the anomoly we see in complex systems? how sure are you of your answer?

    3. 2slugbaits

      Rick Stryker Even though your reply was long and detailed, I think you slid past the important points. First, the so-called “hiatus” is not a climate science puzzle that can’t somehow be explained. Quite the contrary. The “hiatus” is, if anything, overdetermined. For example, a number of years ago Hansen argued that faster than expected ice melt would slow down the growth in observed temperature for two reasons. Ice breaking off and going into the ocean would cool the waters. And heat would be absorbed as ice underwent a phase change. Remember that old middle school science exercise? And in 2005 the National Institute of Standards and Technology predicted a temporary lull in the growth rate of observed temperatures lasting until 2040. This was based on a 64.7 year underlying oceanic cycle. But NIST went to great efforts to argue that this lull in the growth rate was temporary and would mask longer run trends. After 2040 all hell breaks loose in the NIST model. And several years ago climate scientists predicted that solar activity was in a cyclical ebb, so temperature growth would fade for a few years. None of this means that the “hiatus” implies cooler temperatures, only a temporary slowing of the growth rate. So when you read the IPCC report, it’s important to keep in mind that the IPCC is not saying they don’t have clue why the growth rate has moderated; but rather, they have too many explanations that are all consistent with the data and with many models. It’s an embarrassment of riches. The underlying forces that drive global warming are still raging just as strong as ever. In fact, you even admitted that CO2 had a deterministic and physics based warming effect. At this point we’re only arguing about when global warming kills us, not whether.

      I think you also misunderstand the purpose of AOGCMs. No one seriously believes that they use AOGCMs to make long range predictions. The only reason they make predictions is for the purpose of calibrating the models. And they calibrate the models in order to better understand the dynamics between the variables. It’s not even possible to model the weather two weeks out nevermind the next decade. The point of the models isn’t to produce predictions; the point is to try and calibrate the model to better understand the internal dynamics. We do the same thing in economics. Some models are lousy at forecasting, but very good at explaining the dynamics between variables. Other models are good at predicting and completely ad hoc in terms of internal dynamics. If you want to predict the climate 100-200 years out, I wouldn’t advise using an AOGCM because over the very long run you don’t need to worry about a lot of noise. Over the very long run only a very few physical forces will determine the result. Just like your random walk with drift. Over the long run the random walk vanishes and the drift term dominates. Those basic few strong forces will overwhelm the noise that an AOGCM tries to model.

      An AOGCM is a coupled system of differential equations with interaction terms. In other words, analytic solutions are impossible. The best you can do is to approximate the system as a set of linear equations and then decouple. Or if that fails, draw pictures to get the gist of what is happening. In any event, they are computational hogs. And even after all that effort, at the end of the day you’re still left with a modeling approximation of reality and not reality itself.

      You said: since global warming alarmism depends on the correctness of the models, there is no reason to put much trust in global warming alarmism. We should all relax. There is no emergency.

      This is completely, 100% wrong. For purposes of decision/risk analysis by policy makers, global warming models only have to be plausible scenarios with some non-trivial probability of occurring. A 5% chance of a 10 degree Celsius warming by 2200 basically ends the argument. Even if there’s a 95% chance that things will be okay, the policy choice is dominated by the small probability of an unacceptably awful outcome. This isn’t like a normal investment where you live to invest another day if things go south. You should not relax. Your hair should be on fire. There is an emergency. It is entirely plausible that what happens over the next 30 years completely determines what happens 200 years from now. We’re talking about huge inertial forces once they get going. Yes, at the very longest we might have 50 years rather than 30 years. Or we might already be past the threshold. We don’t know with certainty. But we don’t have more than 50 years and it will take at least that long to reorient the economy. I think you’re trying to look for any flimsy excuse you can to avoid any personal inconvenience that might come with an earnest global warming policy. That was the point of Menzie’s cartoon.

      It’s interesting that almost all conservatives fully understood and agreed with the right way to handle asymmetric risks with unacceptable costs during the Cold War. And we accepted this approach even though we did not understand the USSR or deterrence theory or the costs of nuclear nearly as well as we understand global warming. I think the difference is that the Cold War directly impacted people who were alive at the time. In other words, in their heart of hearts (if they have a heart), conservatives just don’t care about future generations. Yes, they care about their own particular kids, but not future generations in the abstract.

      1. CoRev

        2slugs, I have stopped reading most of your comments as they are usually so very wrong, but this one had me laughing out loud. I won’t take it apart because that would embarrass you, but the comment is the brightest you have used, maybe ever. “We do the same thing in economics. Some models are lousy at forecasting, but very good at explaining the dynamics between variables. Other models are good at predicting and completely ad hoc in terms of internal dynamics.”

        You just explained why Climate and Economics are such poor Sciences. Reliance on models that don’t model, and models that can predict but we don’t know why is your answer to why the GCMs are so bad?!?

        Can any one spot the logic error, cognitive dissonance, and blind belief in 21slugs statement? “At this point we’re only arguing about when global warming kills us, not whether.

        I think you also misunderstand the purpose of AOGCMs. No one seriously believes that they use AOGCMs to make long range predictions.

      2. aaron

        “A 5% chance of a 10 degree Celsius warming by 2200 basically ends the argument. Even if there’s a 95% chance that things will be okay, the policy choice is dominated by the small probability of an unacceptably awful outcome. ”

        I think I might shoot myself in the head. I can’t stand the stupid anymore.

  13. Ricardo

    Rick Stryker,

    I have given up on Menzie ever understanding Bastiat or Henry Hazlitt. Good luck my friend!

  14. baffling

    i do find it ironic how deniers argue that 100 year time sample of rising temperatures is too short to justify global warming, but 15 years of stable temperature is more than adequate to say global warming does not exist with certainty!

    1. CoRev

      Baffled, i noticed you have switched tactics to snark and personal attacks. Does that help you in any way?

      1. baffling

        nothing snarky about my comments. justify why a 15 year lull proves your argument against global warming but a 100+ year trend is hocus? then again, i am conversing with somebody who was unable to understand how to estimate the derivative over a time period and interpret the results. math is not your strong suit.

          1. baffling

            no snark. a perfect description of your skill set-no math skills. snark would be describing your math skills as the equivalent of an abacus.

        1. aaron

          I don’t get your argument. The hiatus doesn’t tell us there is no ghe, it tell us it is not important enough to deal with on the emissions end.

  15. baffling

    corev and rick stryker, just curious about your views.
    do you believe in evolution, or should we teach alternatives theories such as creationism and intelligent design in our science curriculum?

  16. JL

    May I recommend to consider that economics is not a science.

    And for the man made climate thing, our climate depends mostly on the sun’s activity, men could help by burning more fossil fuel for doing good.

  17. Menzie Chinn Post author

    Alex: I think I proffered an example of our sun going nova, not supernova. The latter are apparently a more rare event.
    Rick Stryker: Let me re-quote you exactly, word-for-word:

    You see, in science, you don’t prove the theory by showing that it describes the existing observations. You prove the theory by showing that it predicts data that haven’t been observed.

    In your rebuttal, you then go on a long exegesis about in-sample vs. out-of-sample prediction; yet your original comment argues quite clearly that one can learn nothing from in-sample fit.

    Well, as someone who has worked long on exchange rate modeling and forecasting, I think you have, and are, overstating the case. If nothing at all is to be gained from in-sample prediction, then we should throw away all Classical inference. Why do I bother teaching the Gauss-Markov theorem? Taking your point literally, I should posit no point in testing that rate of acceleration on earth is 10m/s2 — someday there might be an observation where, instead of an apple falling, it rises, and accelerates at X m/s2 toward a geosynchronous orbit (the gravitational equivalent of creationism, perhaps?).

    If you want to say that the long time spans involved in out-of-sample testing of climate models complicates matters — sure, I understand complications. But you can’t run away from your absolutist position that nothing can be gained from in-sample inference — just like you can’t squirm away from your previous position that Mitt Romney’s characterization of the 500K/mo job creation rate was perfectly reasonable. If you want, I will dig up each and every comment you made that argued that first, 500K was reasonable if expressed in percent terms (falsified), and then reasonable if you extended the sample back to 1947. At those junctures, you were explicitly defending those figures as representations of the contemporary U.S. economy — not merely “aspirational.” So if you want to re-hash, I am happy to devote another, new, blogpost to reviewing our exchange on the topic.

    Oh, by the way, I’d like to see your dissection of the paper coauthored by Jim Stock, discussed in the post that you commented on and provided the basis for the current post. The analysis is a pretty standard cointegration methodology; is zero gained from the analysis? If so, I think you should pass on your views to Professor Stock, under your own name, instead of a pseudonym, if you truly have the courage of your convictions.

    You also might want to update your views on out-of-sample forecasting; Meese-Rogoff was written over 30 years ago. One interesting discussion is here.

    1. CoRev

      Menzie,. let me make very short and quick analysis of the Stock et al paper, out of date and incomplete! Knowledge has moved on from when it was written. It is another model-based study which relies on climate “forcings” and assumes CO2 as the driver of temperature change. After over 100 comments many of which called out the shortcomings of the even more modern climate models you seem to claim this paper superior? Its introduction starts: ” Evidence for the effect of human activity on climate comes from two sources: experiments run by climate models and statistical analyses of historical data.” Models are NOT EXPERIMENTS. They are implementations of best guesses of the available science, and that misconception pervades the climate science community. Models are NOT EXPERIMENTS, and they do not replicate temperatures even over the short term, and even today’s models can not replicate many of the know forcings.

      Please, understand the importance of the hiatus in falsifying the models. Also understand that the models fail primarily with their interpretations of the forcings. This paper is deficit in defining the known forcings of the time as well those currently identified and their associated relative impacts.

      This analysis is a very short and clear description of what the models are doing: http://wattsupwiththat.com/2013/12/01/mechanical-models/ It concludes: “The way that I am modeling the models is to use a simple lagging of the effects of Equation 1. The equation used is:

      ∆T = lambda * ∆F * ( 1-e^( -1/tau )) + ( T[n-1] – T[n-2] ) * e^(-1/tau) [Eqn. 2]

      In Equation 2, T is temperature (°C), n is time (years), ∆T is T[n] – T[n-1], lambda is the sensitivity (°C / W/m^2), ∆F is the change in forcing F[n] – F[n-1] (W/m2), and tau is the time constant (years) for the lag in the system.”
      and his results were:
      “In all cases, the use of Equation 2 on the model forcings and temperatures results in a very accurate, faithful match to the model temperature output. Note that the worst r^2 of the group is 0.94, and the median r^2 is 0.99. In other words, no matter what each of the models is actually doing internally, functionally they are all just lagging and resizing the inputs.”

      Menzie, with this analysis in mind, and my knowing the models used by your referenced paper were deficient in several ways in the use of the forcings, I can only conclude the findings in the paper are incorrect because its models were out of date and incomplete! Doesn’t matter the credentials of the authors it has not stood the test of time.

      1. Menzie Chinn Post author

        CoRev: Reconciling anthropogenic climate change with
        observed temperature 1998–2008, PNAS (2011)
        :

        Given the widely noted increase in the warming effects of rising greenhouse gas concentrations, it has been unclear why global surface temperatures did not rise between 1998 and 2008. We find that this hiatus in warming coincides with a period of little increase in the sum of anthropogenic and natural forcings. Declining solar insolation as part of a normal eleven-year cycle, and a cyclical
        change from an El Nino to a La Nina dominate our measure of anthropogenic effects because rapid growth in short-lived sulfur
        emissions partially offsets rising greenhouse gas concentrations. As such, we find that recent global temperature records are consistent
        with the existing understanding of the relationship among global surface temperature, internal variability, and radiative forcing, which includes anthropogenic factors with well known warming and cooling effects.

        Which is what 2slugbaits and others commenting on this thread have been saying, and you have been ignoring.

        1. CoRev

          Menzie, how does a conflicting statement refute in any way my analysis of the paper?

          Do you understand what they are saying in your quote? Translation of sentence 1 Although CO2 has risen there is an unexplained hiatus. Sentence 2 – We find the hiatus and a reduction in forcings coincident. Sentence 3 – In our generalized list of (only) three forcings we find reduced (remember reduction in last sentence) Solar radiation, ENSO and aerosols dominate (these forcings are all or predominantly natural) the warming from AGHGs. [Reduced forcings overwhelm increasing AGHGs?] Current temperature records showing a hiatus are consistent with our understanding of climate science.

          Except this paper was written using the previous version of the models and we already know how they have performed. Poorly. You and the “believer” cohort have ignored or argued against the Spencer analysis of model failed performance. With the analysis in your quote they say that temperature change resulting in a hiatus is dominated by natural forcings not anthropogenic GHGs. They ignore forcings influence on rising temps, but you should be able to guess.

          Finally, this paper does not include many of the latest forcings, so my analysis of your paper stands. It was wrong when written, and time has only refined how wrong.

          Its now your turn. How do you interpret your own quote and how does it apply to my analysis. Also, how about the analysis of my reference of the simple “Black Box” analysis of the models. If you accept that analysis, then explain how an incomplete list of forcings effects your paper’s analysis.

          1. CoRev

            Let me take this opportunity to add in reference to your comment: “Which is what 2slugbaits and others commenting on this thread have been saying, and you have been ignoring.” Yes, I often ignore 2slugs commenting due to histoo often lack of logic, arrogance and just flat being wrong.

            Accordingly, I went back this AM and read his comment in this thread and was literally laughing out loud at it. What did bother me is you took it as supporting some position(s). Do you actually agree with his statement re: economics models? ““We do the same thing in economics. Some models are lousy at forecasting, but very good at explaining the dynamics between variables. Other models are good at predicting and completely ad hoc in terms of internal dynamics.”

  18. Synapsid

    Two points:

    The Sun will not go nova, nor will it go supernova. It will become a white dwarf and cool and cool and… A star needs a companion star to have a chance to go nova; only stars several times as massive as the Sun can go supernova.

    Theories are not proved in science–there is no such thing as scientific proof. The goal is to test them to see if they can be disproved. We can’t prove the Sun will rise tomorrow, but we can show very clearly that someone would have to be deliberately obtuse to refuse to agree that it will. Look in the peer-reviewed science literature (the journals Nature and Science are a good place to start) and see if you can find reference to a theory being proven.

  19. Synapsid

    Correction to my previous comment:

    The Sun will go neither nova nor Type I supernova; a companion star would be needed for either to be possible, and the Sun has none. The Sun will not go Type II supernova because it is not massive enough.

  20. anon and lazy

    Menzies silliest post yet… We HAVE the data and the predictions from climate models are consistently wrong. All the DATA tell us there was a medieval warming period followed shortly by the little ice age, and yet some (of the more prominent) global warming alarmist models told us there was no medieval warming period and no little ice age. And the grant-seeking alarmists used this faux science to scaremonger even more …….. until their fraudulent shtick was exposed by McIntyre and McKitrick. Looking at the temperature data from 1980 forward, as shown in the WSJ (see link posted above by Manfred @ 12:14) , we see that these so-called climate models over predict temperatures EVERY year, and the errors are huge. So, here are models where the errors (predicted temperature minus actual temperature) are ALWAYS positive, and Menzie tells us we should have faith in the model’s predictions. I guess he’s gonna tell this to his econometrics class too. This isn’t science, it’s junk.

  21. Menzie Chinn Post author

    Synapsid: Yes, but as I point out, your theory cannot be proven until we get the data…at least using the gospel according to Rick Stryker.

    1. Synapsid

      Menzie,

      What theory?

      No theory can be proven in the natural sciences–that was my statement. You say “your theory cannot be proven until…” What are you talking about?

      There is no proof of theories in the natural sciences. Theories are tested to see if they can be disproven.

      1. Menzie Chinn Post author

        Synapsid: We concur; I was pointing out that using Rick Stryker’s criteria, we would need to wait, …, and wait…, and wait…

  22. don

    JUST BECAUSE A MODEL DESCRIBES THE EXISTING DATA DOESN’T MEAN THAT IT WILL DESCRIBE DATA THAT HAS NOT BEEN OBSERVED

    How many researchers try multiple equations and different variables to justify a hypothesis, forgetting that each variation costs them a degree of freedom? That is an important lesson from the quote.

  23. Menzie Chinn Post author

    don: Well, I think everybody who’s read Ed Leamer’s 1980′s “Let’s take the con out of econometrics” knows that point. I read it in the “basic” econometrics course in grad school.

  24. Rick Stryker

    Menzie,

    No, I did not go into a long exegesis on in sample vs. out of sample prediction. Instead, I explained in some detail exactly what I meant and what I feel is wrong with the climate model argument, quoting passages from the IPCC document to support my claims. You pass over that in silence.

    No, I did not imply that you learn nothing from an in sample fit. That’s a ridiculous characterization. I’m talking about Physics here, which should be obvious. I’ll give yet another example so that there should be no misunderstanding. Newton’s theory of gravity has been remarkably successful, explaining the motion of the planets, the moon, tides, etc. It is confirmed by much experimental evidence. The general theory of relativity is also a theory of gravity. General Relativity says that gravity is nothing more than the curvature of space-time. which is radically different theory from Newton’s theory. When general relativity was formulated, it explained the same known observations as Newton’s theory. Einstein showed also that his theory could explain a known observation that Newton’s theory couldn’t–the perihelion of Mercury. But the acid test is for the theory to predict an effect that has not been observed. General Relativity predicted that light would be bent around the sun and Eddington made detailed observations during a solar eclipse that confirmed the theory. At that point, general relativity became the accepted theory of gravity.

    Global climate models are just complex physical theories–basically computational fluid dynamics. They should also make predictions about unobserved phenomena which are confirmed by experiment in order to be taken seriously. Explaining what’s already known is not good enough.

    Just to get into the details of global climate models a bit more, I already explained the problem with tuning parameters. But there are couple more major problems with these models.

    The first problem is that the behavior of clouds is critical to what happens to the temperature but clouds cannot be modeled precisely. Clouds can generate positive or negative feedback effects. Low level clouds reflect sunlight into space and so have negative feedback, lowering the temperature. High level clouds reflect radiation from the earth and so have positive feedback. And thick clouds do both, having neutral feedback. The details of the cloud’s reflectivity also matters a lot. Aerosols can produce positive or negative feedback directly. They can also act on feedback indirectly, by enhanceing cloud reflectivity, which can produce positive or negative feedback, depending on the type of cloud. You really have to get the clouds right in these models. Depending on what actually is happening with the clouds, you could get big or little temperature effects.

    The second problem is that biology is very important too. Huge amounts of carbon are transferred into and out of the atmosphere by plants. How much is really known about this process, especially if the temperature or level of Co2 changes? And yet climate models are primarily physical theories, not biological theories.

    I mention these details to show how complex it all really is. Just because you tune some parameters, run a fluid dynamics simulation, and match the existing temperature record does not mean you know what’s really going on. However, requiring the climate models to predict hithero unobserved phenomena will show whether they have really described the environment correctly.

    I know you’d like to change the subject to my defense of Romney’s remark, exchange rate models, or Stock’s paper. But let’s stick to the question at hand. I’ve defended my assertion with detailed and specific arguments about the climate models, which you’ve not addressed. So far the only argument you’ve made is to post a cartoon.

  25. Rick Stryker

    Corev,

    Thanks. I appreciate the comments you are making as well.

    I agree that the state of climate research is a great case study for group think, funding impacts, etc. But there are other prominent examples.

    For example, string theory, which attempts to unify general relativity and quantum mechanics, completely dominates theoretical physics. It’s beautiful theory and beautiful mathematics. There is just one problem. There is no experimental evidence. String theory explains everything known but it has not passed the test of predicting correctly some new phenomena. And it doesn’t appear that such a test is feasible now or in the foreseeable future. For that reason, I know quite a lot of theoretical physicists who have gotten out of physics and entered other fields, such as biology or finance. For people who have left, the feeling is that string theory is not really science, since it can’t make testable predictions about unobserved phenomena. But if you stay in physics, it’s pretty much impossible to challenge it. So, it’s a similar situation to climate science. People inside theoretical physics will say there is a consensus. There is, since people who disagree must remain silent or get out. People who stay in it believe in it strongly and think the experimental confirmation will come. If someone does figure out how to test it some day, we’ll know the truth. And if the theory pans out, the current string theorists will be heroes. Otherwise, the whole episode will be a massive case study in group think, perverse incentives, etc.

  26. Menzie Chinn Post author

    Rick Stryker: This is a futile conversation. My point remains that your absolutist statement: “You see, in science, you don’t prove the theory by showing that it describes the existing observations. You prove the theory by showing that it predicts data that haven’t been observed.” cannot be defended. You have now backed away from that strong assertion by saying you “did not imply that you learn nothing from an in sample fit.”
    All the rest of your exegesis is irrelevant once we see you try to squirm away from that original assertion. Now you say that predicting the unobserved is “the acid test”. OK, but that’s your view.
    Finally, I’m happy to allow that complete verification of global climate models will wait a long time, if ever. Of course, this is true of almost every economic model — consider the expectations hypothesis of the term structure, or uncovered interest parity, or the present value model of stock prices — the important variables are unobservable. So we are forced to rely upon joint hypotheses, in-sample testing and the like. And yet, I suspect you wouldn’t say there’d been zero progress in science. Or maybe you do.
    Anyway, for my tastes, I’ll stick to (1) formulation of hypothesis, (2) testing, (3) interpetation as my interpretation of the scientific method.
    I won’t mention 500K jobs/mo if you stop trying to re-write your defense of that figure as a description of reality to an “aspirational” goal.

    1. CoRev

      Menzie, your argument against Stryker is getting weaker. The subject in which his comment was made was the value of the GCMs in their ability to predict. If they fail predicting they fail to fully understand the issue and the science is either wrong of incomplete. You then hang your argument on the value of models in analyzing their internals?!? Even he admitted this could be of value. Neither admit the value of improving the internals and the uncertainty associated with that new knowledge when the overall model fails to recreate the out sample, the prediction. So the model does not model, and that is OK? You actually can say this with a straight face? ” Now you say that predicting the unobserved is “the acid test”. OK, but that’s your view.” Do you actually think a model that doesn’t predict can be used to make policy decisions that cost trillions, to change what 2slugs thinks they predict:” when global warming kills us, not whether”?

      You admit, and I am surprised: ” I’m happy to allow that complete verification of global climate models will wait a long time, if ever. Of course, this is true of almost every economic model — …( a list of variables here) — the important variables are unobservable.” Unobservable, or too little is known about them to model?

      Because of your explanation of the similarity of the modeling, I better understand how badly economics models relate to reality. And why you would consider clarification of the model variables is the answer and not their overall ability to represent reality. If that’s all you have then you go with it, but dont react/over react when your policy recommendations costing trillions or the interpretations likes 2slugs (and yours?) “when global warming kills us, not whether” are questioned.

      Stryker is far more correct than you are!

  27. Rick Stryker

    2slugbaits,

    You have a very serious misconception if you think that the global warming predictions don’t come from AOGCM models, as you claimed.

    I noted with amusement your argument that even a 5% chance that global warming alarmism is true justifies drastic action. That sure simplifies the problem. We don’t have to worry about the facts anymore. According to you, even if the probability that a massive temperature increase will occur is small, the consequences of such an increase are so bad for humanity that we are compelled to take action. You are making the Pascal’s wager argument for global warming alarmism, which many people resort to when they are having trouble defending it.

    For those unfamiliar with the argument, Pascal argued for the existence of God as follows: I have very little evidence for the existence of God. What should I do? If I choose to believe in him and he doesn’t actually exist, then I give up the fun life I could have had as a sinner. But if I don’t choose to believe in him and he does exist, I suffer eternal torment. Since eternal torment is really bad, even a very small chance that God exists compels me to believe in him.

    Translated into global warming alarmism, the argument goes: I have very little evidence that the temperature really will rise significantly. What should I do? If I choose to believe in global warming and the temperature doesn’t rise, then I give up some economic growth from the higher taxes and regulations that were imposed on carbon consumption . But if I’m a liberal it’s not all bad since I get all that carbon tax money to spend plus regulations to implement and supervise. On the other hand, if I don’t choose to believe in global warming and the temperature does rise, I’m in environmental Hell. Therefore, Al Gore just needs to convince me that there is a 5% chance that what he’s saying is actually true. That’s enough evidence for me to go along completely with his liberal agenda.

    I wonder if this sort of argument really works on someone who isn’t secretly lusting for all that carbon tax money, those new regulations, and all the commissions, committees, and studies by academic experts? I really doubt it.

  28. Rick Stryker

    Menzie,

    I didn’t back away one iota from my assertion and I strongly defended what you call “indefensible” by giving explicit examples how it is true. My first example went through the details of how global climate models work and explained why you need to make predictions on unobserved data to prove the theory. I followed up with additional reasons for the global climate models in my next comment. My second example was from the history of science where I showed that scientists, not Rick Stryker, required that General Relativity make a correct prediction about some unobserved phenomena before being accepted. That’s not an anomalous requirement but a general requirement in the natural sciences. In my comment to Corev, I noted that many physicists have left string theory because they don’t think it meets the requirement. I can only think that your continued failure to understand this point reflects your background in economics, which has a different methodology. You still have not refuted any of these arguments in any way other than to claim that it is unnecessary to refute them.

  29. Ecomedian

    “Curve fitting” of models to data using unmeasurable free parameters is a well-known fallacy in the “hard sciences”, yet seems to be pretty central to certain modeling in economics. Noah Smith made this point a couple weeks ago with his “Skippy the Flying Mongoose” free parameter model. NAIRU is an example of a parameter that can fit any model to data–if inflation data goes against you, the unmeasurable NAIRU can “respond” and inflect your model’s “prediction” to accommodate.

    Theories that make predictions about experimental results, and get validated by “out of sample” data without the benefit of retroactive parameter tuning are preferred in the scientific method.

    How did Menzie’s published model a year ago, based on historical data, that predicted export growth benefit from Abenomics, fare against future data? Got J-curve?

  30. Rick Stryker

    Menzie,

    As I thought about this, even your economics background does not explain your refusal to see this point. Allow me quote from Milton Friedman’s “Essays in Positive Economics.”

    “The ultimate goal of a positive science is the development of theory” or “hypothesis” that yields valid and meaningful (i.e., not truistic) predictions about phenomena not yet observed. ”

    Not yet observed. Did you get that? That’s the first sentence of the section “Positive Economics.” Now compare to my statement:

    You see, in science, you don’t prove the theory by showing that it describes the existing observations. You prove the theory by showing that it predicts data that haven’t been observed.

    Now let’s go on to get another quote from Friedman:

    “To avoid confusion, it should perhaps be noted explicitly that the “predictions” by which the validity of a hypothesis is tested need not be about phenomena that have not yet occurred, that is, need not be forecasts of future events; they may be about phenomena that have occurred but observations on which have not yet been made or are not known to the person making the prediction. For example, a hypothesis may imply that such and must have happened in 1906, given some other known circumstances. If a search of the records reveals that such and such did happen, the prediction is confirmed; if it reveals that such and
    such did not happen, the prediction is contradicted. ”

    Is Milton Friedman’s statement absolutist and indefensible as well?

  31. CoRev

    Thanks! Menzie’s rope-a-dope is failing with too many body blows. The laughter from the masses and even some liberal papers at Kerry’s and Obama’s latest comments re: climate change show us that the tide has significantly turned against it. Ole Mom Nature always has the last word against human hubris. I dunno, though, if that’s enough against some individuals’ hubris.

  32. baffling

    rick stryker,
    we have a conservative block who insists on spending trillions of dollars on “defense”, first in the cold war and then against the terrorists. the cold war was a low probability event of escalation. and yet it was supported to a very large degree, why? because the outcome of that low probability event was unacceptable. it is baffling how the same people who supported the cold war are against global warming, considering the similarities in outcomes. other than ideology.

  33. baffling

    since the question was ignored, i will ask it again;
    “corev and rick stryker, just curious about your views.
    do you believe in evolution, or should we teach alternatives theories such as creationism and intelligent design in our science curriculum?”

  34. Menzie Chinn Post author

    Rick Stryker: But Milton Friedman, in his analysis of the the permanent income hypothesis, did what — he used classical hypothesis testing. To the best of my recollection, that was in-sample analysis. I don’t think he said, let’s wait a hundred years to see how it fits.
    All your long-winded description of the global climate change model told me is that they share similarities with many economic models — including DSGEs; they’re complex (nonlinear), they rely upon combinations of estimated and calibrated parameters that are updated, and ultimate validation will come — perhaps — sometime before the sun burns out.
    I referenced the Kaufmann, Kauppi, Stock paper — not to divert attention — but rather to bring attention to the fact that there were cases where the analysis was conduced using statistical methods that we economists (or at least I) could understand, that yielded information from in-sample data. That was the counter to your “You prove a theory…” — and contra Friedman. But then Friedman’s works are not received wisdom — I still await acceleration toward negative infinity inflation.
    CoRev: When you find a time series on the market’s expectation of inflation (rather than a proxy), send it along. I can’t wait to publish the paper that will break such new ground!!! (Your statement indicates a complete and utter lack of comprehension of modern macroeconomics and finance — would it kill you to read a textbook on economics if you are going to comment on economic models?).

    1. CoRev

      Menzie, let me remind you what I said was a quote from your very own commenter 2slugs. Do you agree with him or not?

      You’re playing with a subject with huge economic and political ramifications, and the views outside the “liberal bubble” do not coincide with those inside the bubble not the faculty lounge. Moreover, you are dealing with adults who have spent years following and studying the subject during a time frame where views on the subject are changing and solidifying.

      Get over yourself. On this subject you are a novice as are Kerry, Obama and his science crew. Worse for them, that knowledge is growing amongsth the unwashed voters who care.

  35. 2slugbaits

    CoRev Do you actually agree with his statement re: economics models? ““We do the same thing in economics. Some models are lousy at forecasting, but very good at explaining the dynamics between variables. Other models are good at predicting and completely ad hoc in terms of internal dynamics.”

    Ever read a book on econometric models? Are you familiar with overparameterized VARs? Absolutely lousy at answering some questions and excellent at answering others. Another example. I would never want to use a Solow or Ramsey neoclassical growth model to predict next year’s GDP. Short run dynamics are usually best modeled when you focus on demand shocks. In some contexts an NK model. In other cases (like liquidity traps), an old school IS-LM with all of its ad hoc assumptions. In other cases you might want to throw theory out the window and used an unstructured VAR. I don’t think you have ever actually had any real world experience with models, so perhaps that’s why don’t understand that some models are better at forecasting and others are better at teasing out internal dynamics. Here’s a good primer that makes the point…and I’ve used examples from this text when I’ve taught this stuff…for example, with Dr. Roy Spencer’s grad students a few years ago.

    http://www.wiley.com/WileyCDA/WileyTitle/productCd-EHEP000338.html

    And no, climate scientists do not make long run predictions based on AOGCMs. The predictions that they make are for the purpose of calibration and intended for internal consumption. Look at any textbook on climate science. There will be a chapter on CO2 and greenhouse gases. There will be a chapter on ocean currents. There will be a chapter on winds. Each chapter will present a model that more or less stands in isolation from the models in the other chapters. Academic climate modelers try to integrate those models because they want to understand the internal dynamics, not because they think they can offer long run projections. You see the same kind of thing with economic models. No one that cranks out an NK-DSGE model thinks that it can predict the economy 10 years out.

    Rick Stryker I think you botched the Pascal story. The reason Pascal’s argument was wrong was because he forgot that you couldn’t fool God. In any event, I did not say that we should fight global warming even if there was no evidence for it. I said all that was required was plausible evidence with a nontrivial probability. There is no plausible case for God’s existence, so the two examples are not at all comparable. And I really don’t think you believe your own argument anyway. Or at least you don’t believe it when the logic of asymmetric risk is applied to other aspects of life. I’m guessing you wouldn’t jump from a plane if there was a 5% chance that the parachute might not open in exchange for a 2% increase in income. But that’s very close to the trade-off with global warming. Except…the main difference is that it wouldn’t be you jumping out of the plane; it would be some future generation. And you’re quite comfortable putting that future generation at great risk with an unreliable parachute if it means greater income for you. Do you really think conservatives during the Cold War carefully balanced the failure of deterrence theory against the cost of the Cold War??? Seriously? The political and military models that guided us during the Cold War were far more uncertain than the theory underlying global warming, but that didn’t stop conservatives from wanting us to spend two or three times what it would cost to fight global warming.

    1. CoRev

      There I ignored your comment then went back for the laugh. You have been repeatedly wrong on this claim: “And no, climate scientists do not make long run predictions based on AOGCMs. ” But, the IPCC says this: “Each IPCC assessment has used new projections of future climate change that have become more detailed as the models have become more advanced. Similarly, the scenarios used in the IPCC assessments have themselves changed over time to reflect the state of knowledge. The range of climate projections from model results provided and assessed in the first IPCC assessment in 1990 to those in the 2007 AR4 provides an opportunity to compare the projections with the actually observed changes, thereby examining the deviations of the projections from the …”

      Before we get into a discussion of the projections’ lengths, the IPCC says this: “Projections for the next few years and decades are sensitive to emissions of short-lived compounds such as aerosols and methane. More distant projections, however, are more sensitive to alternative scenarios around long-lived GHG emissions. These scenario-dependent uncertainties will not be reduced by improvements in climate science, and will become the dominant uncertainty in projections over longer timescales (e.g., 2100) (FAQ 1.1, Figure 1).” In the quote you will find a reference to accompanying graphs. If you have ever read an IPCC report it is common to see graphs of climate change projections to 2100, and nearly a full century is a long term in almost anyone’s book except 2slugs and his text books.

      Does anyone actually believe he has read a Climate Change text book?

      I also see you doubled down on the poor quality of economic modeling. I accept your own view (and I presume Menzie’s) that the models can not model a full economy.

      BTW, you are wrong about your assumption: ” I don’t think you have ever actually had any real world experience with models,…” In fact I helped define and automate several physical process models. And, then became an accomplished user of them for predictions. Knowing which dials to turn can be helpful.

  36. baffling

    rick stryker
    ““The ultimate goal of a positive science is the development of theory” or “hypothesis” that yields valid and meaningful (i.e., not truistic) predictions about phenomena not yet observed.
    Not yet observed. Did you get that? That’s the first sentence of the section “Positive Economics.” Now compare to my statement:
    You see, in science, you don’t prove the theory by showing that it describes the existing observations. You prove the theory by showing that it predicts data that haven’t been observed.”

    actually you are simply bastardizing the understanding of theory in science for your own argument. theory is developed to better understand observations made today that do not exist in the current theoretical framework. their predictive capabilities will come through in the future to confirm or deny the generality of the theory. but typically the theory is developed to gain better insight into already made observations. that is what is happening with global warming models. yes we want and need predictive capabilities, but we also need better understanding of parameters we can influence today. it does you no good to wait 200 years to confirm, yes excess carbon emissions are bad. great you finally verified your prediction-too bad the continental shorelines have receded by a mile around the globe.

    a perfect example of theory in action is quantum mechanics. it was developed initially to explain observed behavior that did not match classical theory. its predictive powers were only accepted after verification of past data points. we continue to model and upgrade global warming along the same lines. i think you fail to understand some important distinctions between a theory of the natural world and a theory from economics-they are really not the same. the laws of physics are far more demanding than “laws” of economics.

  37. baffling

    corev and rick stryker, since we are debating the legitimacy of science in these threads, i think readers would really like to know where you stand on the topic
    “do you believe in evolution, or should we teach alternatives theories such as creationism and intelligent design in our science curriculum?”

    just curious where a denier of global warming stands on the scientific merits of evolution. the silence is baffling.

    1. CoRev

      Yup, I get asked ofter: ” i think readers would really like to know where you stand on the topic .. evolution…” And, its almost always from someone who has gone beyond their meager knowledge base in an extended debate. Debate the points we have made. There certainly is much meat there to be torn at, but changing the subject to something unrelated is a silly and useless trick.

      1. Nick G

        Actually, it’s completely related: I think you’ll have a hard time finding any professional climatologists, paleontologists, geologists or biologists who doubt evolution. Really, any branch of science that relates to long-term historical events.

        Climatology is relatively new – not like astronomy or biology – but it seems to provide a new intellectual litmus test. In the past, one could identify someone who was unable to rise above their group’s beliefs to think scientifically by asking them if they thought the earth was round, or if evolution was valid. Now, one can identify them by their inability to agree that Climate Change is a serious problem.

        This is related to the fact that anti-Climate Change arguments are just part of group/follower thinking. How can we tell?

        First, because Climate Change really is the scientific consensus, and it’s the international consensus: even oil exporters like Russia and Saudi Arabia concede that it’s valid. China concedes it too, and they’re working harder on it than the US. China has more than enough scientific expertise to independently review climate science, and plenty of motive to disagree, yet they don’t.

        2nd, there’s a really obvious difference between political parties in the US: the Democrats follow the scientific and world consensus, and Republican’s don’t. There aren’t any leaders of the Republican party that are willing to support Climate Change.

        3rd, the republican rank and file generally agree with their leadership. Now, you can’t really argue that those republican followers have all read all of the Climate Change science and come to an independent opinion – it’s quite obvious that most of those who hold anti-Climate Change opinions are doing so because of they identify as part of a group that holds that idea. There’s a striking correlation with other anti-science ideas as well: most republicans don’t believe in evolution.

        So, ask yourself, Do I hold this idea just because I belong to a group that believes it? Do *any* of my friends disagree with me? Do any of the websites I like to read disagree with me?

        1. CoRev

          NickG, I think it very important that you ask yourself your questions: “So, ask yourself, Do I hold this idea just because I belong to a group that believes it? Do *any* of my friends disagree with me? Do any of the websites I like to read disagree with me?” On this subject I personally read sites with both views to get a wider understanding of both the science and the arguments. Can you say the same? Do you actually go to many CC sites? If you do it is not evident from your commenting.

          The biggest hole in your and baffles claims about how others/conservatives think, feel, and are passionate about is you generalize to individuals. Those individuals intuitively know how close you are to their actual thoughts,feeling and passions. Accordingly they discount you because your evaluation(s) are so far from what they actually are. In a short answer, you don’t know me, and you are not even close in describing me.

          Finally on the subject at hand this weekend, Climate Change, please define just what it is with some precision. As in you personal generalizations you have used extraordinarily general terms: “… identify them by their inability to agree that Climate Change is a serious problem.” While you are defining just what Climate Change is to you define and find historical or scientific references (not models-based because we have shown conclusively that they are not capable) for those serious problems.

          If you do not do this, then you will confirm my own thoughts that you and baffled are the ones buried in group think.

          1. Nick G

            you don’t know me, and you are not even close in describing me.

            Sadly, I know a lot of people (friends, family members, other members of my “tribes”) who say things that are very similar. And, yes, all of them would have a very, very hard time agreeing with what I’m saying. It’s very, very hard to admit that one isn’t being what one imagines to be purely rational – that one is simply following others. People who don’t believe in evolution won’t agree that they’re following someone else’s ideas. People who believe in leprechauns won’t admit it. And yet, it’s obvious. If you don’t believe it about your own group, you can see it in other groups.

            If you identify as a “conservative”, then you probably talk about “liberals” as just being out of touch with reality. If you’re a Christian, then you probably believe the Muslims are just following what they were taught. And, obviously, most people follow the religions and political beliefs of their parents. Most republicans believe in the party line, and most democrats follow their party line. Do all of those party members just happen to have come to reasoned agreement on dozen of different items??

            Heck, you said it yourself: “my own thoughts that you and baffled are the ones buried in group think”. We all agree that almost everyone is following others’ ideas, right?

            Realistically, this is part of the human condition: we’re just not smart enough to re-evaluate *all* of the culture we’re handed. We don’t have the time! No one does. So, we pick and choose what we reconsider, and some of us are luckier, and get better training in how to do that…

            As far as the definition of Climate Change: I’d call it “The complex effects of gases called Greenhouse Gases.”: CO2, methane, etc. That pollution primarily traps heat, but has other important effects like ocean acidification. As a practical matter, I’m mostly thinking about excess CO2. When I say Climate Change, you can substitute Anthropogenic Global Warming, if you like.

            And, no, I’m not a climatologist, not do I pretend to be one. I know a lot about energy and science: enough to be reasonably confident about evaluating claims. In this case, I’m willing to believe the scientific and international consensus.

            Again, China has the competence and motivation to question Climate Change, but they don’t. Instead, they’re doing more than the US (and really, more than Europe) to transition towards tech that reduces emissions. That doesn’t mean they aren’t using massive amounts of coal, or that their CO2 or even their conventional pollution emissions have started to fall, but they’re making a real effort to transition to new tech.

  38. baffling

    corev,
    its not changing the topic in the slightest. you already challenge the science of global warming. is this an anomaly, or do you hold contrarian views in other areas of science such as evolution? too much skepticism tends to negatively impact one’s credibility. we can discuss whether you do not understand derivatives or do not believe in them next.

    1. CoRev

      Continuing to change the subject to me gets you no points. You object because I caught you cherry picking from a chart and are hiding behind a specious argument. Cherry picked? Yup! Used you own definition for the period you were analyzing (it was not defined in your reference) and then approximated (they too were not in defined in your reference) the data for that period to calculate your amounts. So, now you want to claim that anyone should have percieved your exact meaning?????

      BTW, I too have taken and used calculus. That doesn’t help with mind reading though. The only thing that surprised me is you failed to use that big surge in melting, but I guess even in your desperation you realized that would make your cherry picking too obvious. BTW, what happened to those decades in your original claim?

  39. baffling

    corev, again i must state from your comment then you fail to understand the concept of a derivative and inability of read a graph. you could have “cherry picked” any group of numbers on the graph and obtained the results i noted. the main feature is right in the middle of the graph. you asked when such a rate of sea level rise has ever occurred. the graph showed 7,000+ years of rates in the neighborhood of 1.4 meters/century. so you had nearly 1.5 meters of sea rise in 10 decades. i addressed this already but i will point it out again, that is in the time frame of decades.

    and it appears if you have taken calculus, perhaps it is time for a review course on the subject.

  40. Captcha broken

    sorry for the off topic but it looks like captcha is broken on the Economic Implications of “Anthropogenic Climate Change and Extreme Weather”

    I think 5 X two = 10 but captcha disagrees

Comments are closed.