William M. Briggs

Statistician to the Stars!

Page 146 of 693

That “1-in-27 Million Chance That Earth’s Record Hot Streak Is Natural” Is Preposterous

This post is one that has been restored after the hacking. All original comments were lost.

We met a lot of bad statistics over the years, but this one wins the Blue Ribbon With Gold Lace, Free-Beer-For-Life Award of Statistical Putrescence. It is not only not true that the “1-in-27 Million Chance That Earth’s Record Hot Streak Is Natural” probability is correct, as quoted by Mashable and others who ought to know better, but that it’s so far from True that if it were to travel at a thousand miles per hour for two thousand years it would still be just as far from True as the day it started.

And this is so even if you are a leftist progressive Marxist politically correct environmentalist feminist hater of Fox News Elizabeth-Warren-For-President-button-wearing Democrat who would like nothing better than if the Temperatures Of Doom were just around the corner and all agreed that you—yes, you—should put in charge of the World’s Affairs, so saving us all.

It is a mark, and an important one, of how political climate “science” has become that statistics like this are quoted and accepted, and, yes, even rejected, all because people desire man-made apocalyptic the-time-to-act-is-now global warming be true or false.

The reader will pardon my exasperation. Misunderstandings of probability and bad statistics now account for 83.71% (p &lt’ 0.001) of all bad science. I’ve reached the snapping point..

We’ve met this statistic, or its cousin, before, when we discussed why the then touted “More On The 1 in 1.6 Million Heat Wave Chance“, created by somebody at the NCDC was ridiculous. Don’t be lazy. Read that article.

Here is why all these numbers are you-ought-to-have-known-better stupid.

Do you accept that some thing or things caused each month’s or year’s temperature? If not, you’re not familiar with science and so are excused. Well, some thing or things did cause the temperature (however it is operationally defined). Measuring it does not determine those causes, it only records the observations.

Suppose it is true that these causes are analogous to reaching inside a bag of temperatures, pulling one out, and then blanketing the earth with it. Some of these temperatures are high, some low; a mixed bag, as it were. Now let Mother Nature pull out the temperatures and we observe them. What are the chances we see what we see? 100% is the only acceptable answer.

Don’t see that? If it really is true, based on our accepted model, that the chance this year’s (or month’s or whatever) temperature is just as likely to be higher or lower than last year’s, then the probability we see the record we see is 100%. Think of a coin flip. The chance, given the obvious premises about the coin, we see any string of Hs and Ts is the same (HHHHH has the same chance as HTTHT, etc.); thus the chance we see what we see is 100% no matter what we see.

Of course, we may be interested in Hs or High Months more than Ts or Low Months. The chance, accepting our model, of so-many Hs or High Months in the record can also be calculated and will be some number. Imagine some string of High Months then calculate the chance of seeing this many in a row. Make the string as long as you like, which makes the probability weer than wee, vanishingly small. Make the probability smaller than 1 in 27 million. Make it 1 in a billion! Nay, two billion!

Does that mean our model of causation is false? No! No no no no no. No. We accepted the model as true! It is therefore, for these calculations, true. As in True. That means the probability is also true, given this model. But this probability doesn’t say anything about the model, it is a consequence of the model. If you don’t like this, then you shouldn’t have accepted this model as true.

What about that? Why did you accept this preposterous model as true? Who in the wide word of human nuttiness ever claimed that a forgetful Mother Nature regularly reached into a bag of temperatures and cast it over the surface of the deep? I’ll tell you. Nobody.

But that’s just the model everybody who is foisting these silly 1-in-27 Million Chance-like statistics on us believes. Or claims to believe, else they never would have quoted these numbers. But we know they don’t believe the model! So why do they quote these numbers?

Because they want to scare you into believing, without evidence, their alternate model, apocalyptic man-made global warming is true.

Some thing or things caused the temperatures to take the values we observed. If we knew what these causes were, i.e. what the model was, we would state them, yes? That chance of seeing what we saw, given this model, would not be 1 in 27 million, right? Since we would know the causes, the chance of seeing what we project would be 100%. Take an apple and drop it. Given the causal model of gravity, what is the chance apple meets earth? 100%.1

Do we have a full causal model for temperature? We do not. If we did, then meteorologists and climatologists would not make mistaken forecasts. Because they often do (especially climatologists), it must be that their models are incomplete. We do not know all the causes of temperature. But because we do not, it does not—it absolutely does not with liberty bells on—mean that we do know the cause is man-made global warming.

This is the sense that the 1 in 27 million is wrong. It’s the right answer to a question nobody asked based on a model no sane person believes. Its answer is useless utterly in discovering whether global warming is true or false. If this is not now obvious to you, you are lost, lost.

—————————————————————————————

1You are not being clever but obtuse by suggesting that, say, something interferes with the apple’s (and earth’s) path. That interference changes the model, it is an additional premise. The model is no longer gravity, but gravity-with-interference.

More On The Pope And Global Warming

Any non-scientist civilian can be forgiven, if it is not his business to know about the climate, to follow the majority and claim that global warming, a.k.a. climate change, a.k.a. climate disruption, a.k.a. climate chaos, a.k.a. climate of doom, is here, increasing, devastating, and almost entirely caused by man.

The Pope on a plane (not his strongest milieu) was asked about the subject. He was on his way to the PI to chat with those who managed to live through the thoroughly nasty Typhoon Haiyan which, so reported the AP, “the government has said was an example of the extreme weather conditions that global warming has wrought.”

The PI government might have said that, but if they did they were wrong. It is a false statement. Global warming did not sneak into the weather control room, through a switch, and release another typhoon into the wild. Instead, large tropical storms like Haiyan are actually down in number. The typhoon was sure deadly, however. Why?

Well, I lived for three years not too far from the PI on the small island of Okinawa. We got hit by one or two typhoons a year, and these caused little damage, whereas the same storms, or storms of similar magnitude, bulldozed through the PI. On Okinawa, everything is made of rebarred concrete. On the PI there are still many shacks (and that may be a generous term). The solution is obviously to increase the efforts of holding back Nature on the PI as in Okinawa.

The AP also quoted the Holy Father as saying:

“I don’t know if it (human activity) is the only cause [of global warming], but mostly, in great part, it is man who has slapped nature in the face,” he said. “We have in a sense taken over nature.”

“I think we have exploited nature too much,” Francis said, citing deforestation and monoculture. “Thanks be to God that today there are voices, so many people who are speaking out about it.”…

“The [world-government climate] meetings in Peru were nothing much, I was disappointed,” he said. “There was a lack of courage. They stopped at a certain point. We hope that in Paris the representatives will be more courageous going forward.”

Any civilian, I reiterate, whose business is not governance and whose livelihood does not rely (mostly) on matters environmental would say things like this. Civilians are blanketed from birth (since ~1968) with slogans, propaganda, and even the words of honest but mistaken scientists who claim the End Is Nigh. Skeptics, even when joined in voice, can barely manage a whisper. The environmentalists have the bulk of money, resources, and ears.

Yet…our Pope is not a civilian and his business is governance.

Individual men have slapped their environs in the face. Others have caressed. But most of us get along. Yet this isn’t quite right. Something deeper is wrong with the sentiment Pope Francis expressed. Man is part of nature. It is not something separate from us. We’re adapted to live here, not Mars. It is true, necessarily true, that we have manipulated the earth to suit our needs, but that statement is true of every species. It is impossible—as in impossible—for us not to change the climate, the land, the sea.

The only questions are how much should we manipulate, what are the consequences, good or bad, and can we encourage the good and mitigate the bad? Deforestation in developing countries is an understandable response to wanting to feed people, as is monoculture. The solution to both, just like in the PI, is to import the knowledge and technology (and energy, like 100% all-natural organic fossil fuels) that mitigate the maladaptations while keeping the good, which is people with full stomachs.

It’s true that the Peruvian meetings were “nothing much”, but let’s pray the Paris meetings do less. Putting more control of ordinary people’s lives into the hands of fewer and fewer of those curious rapacious monomaniacal Experts we have somehow, inexplicably, let be put in charge of us cannot be a solution. Those would-be Global Leaders stand for every spiritual opposite of the Church. Will the Pope really trade another bike lane in downtown Buenos Aires for accepting gay gmarriage?

The Holy Father’s encyclical won’t be issued until (it is reported) June or July. Which means there is still time for somebody who takes a realistic view of the climate and is close to the Pope to advise him. This isn’t me: I’m a nobody with no contacts. But there must be somebody.

Men Who Post ‘Selfies’ Are Psychopathic, Narcissistic. Science Says So!

Say Cheesy.

Say Cheesy.

This post is one that has been restored after the hacking. All original comments were lost.

Far be it for Yours Truly to disagree with a peer-reviewed study. It must be, Science says, that the gentleman above, all in the habit of playing with selfies (and here, here, here; oh, look the rest up by yourself) are loaded with psychopathic and narcissistic tendencies.

Hey. This isn’t me speaking, I emphasize, it’s Science. Peer-reviewed Science. As found in the paper “The Dark Triad and trait self-objectification as predictors of men’s use and self-presentation behaviors on social networking sites” from the journal Personality and Individual Differences (Volume 76, 2015, pp. 161–165) by Jesse Fox and Margaret C. Rooney, from the School of Communication, Ohio State.

Fox, incidentally, is also author of “The dark side of social networking sites: An exploration of the relational and psychological stressors associated with Facebook use and affordances” in Computers in Human Behavior (Volume 45, April 2015, Pages 168-176). Also “Sexism in online video games: The role of conformity to masculine norms and social dominance orientation” in that same journal (Volume 33, April 2014, Pages 314-320). And several others in the same vein.

Here’s what the ladies did. The went on line and found some 800 men, aged 18-40, and asked them to fill out questionnaires. On a scale of -1.3 to 87, what is your feeling (don’t think) about the validity of conducting science by asking ad hoc questions? Skip it.

What were the questions? The Dark Triadduh dun dunnnn—, Self-Objectification Questionnaire, which asks men “to rank various body traits from most to least important”, and time spent fiddling around on the Internet, especially in posting selfies. The Dark Triad—duh dun dunnnn—asks questions and expects honest numerically valid answers to items like “I tend to want others to pay attention to me”, “I tend to not be too concerned with morality or the morality of my actions”, “I tend to manipulate others to get my way”. That apply to the gentlemen above?

Now the wee p-values—which, I’m guessing, are probably not the trait rated most important by the men. Skip it. Anyway, would you believe it, but the correlation between the Dark Triad’s—duh dun dunnnn—Machiavellianism score and the Trait Self-objectification score was 0.68. Think on it.

Or, rather, don’t.

What statistics-happy researchers do is this (and this topic I’ll greatly expand upon soon). They first ask questions and claim those questions unambiguously and without error measure something, say Psychoticism, and then they ask other questions and claim those unambiguously and without error measure something else, like self-objectification. They then run all manner of statistics to claim that either Psychoticism predicts or explains or causes self-objectification or vicey versey.

This association, “proved” by wee p-values, is never strong, but this weakness is immediately forgotten. Then out comes Theory which says how and why Psychotics self-objectify more than l’homme moyen, or maybe it’s how those who self-objectify might turn out to be psychotic. Everybody forgets it starts with one question in one “scale” or “instrument” being similar (or even the same) to another question in another scale.

And entirely lost is that the whole thing relies on questions with arbitrary numerical scores. As if emotions can be summarized that way.

Well, in a regression model (and, typically, the authors do not understand regression: see this and this) the parameter associated with time claimed spent playing with selfies and the questions in the Dark Triad—duh dun dunnnn—said to represent Narcissism and Psychopathy were sort of wee. Exciting, no?

To describe the procedure and results took just over one page of journal paper. And, as I promised above, just as much space was devoted to Theory and Speculation (under the section called Discussion).

The authors said things like “Those higher in narcissism and psychopathy reported posting selfies more frequently”, which is false. Some who scored high on the Dark Triad—duh dun dunnnn—for these things said they also posted many selfies. But some who did not score high also said they posted many selfies. See what I mean about uncertainty being forgotten?

Theory? “The hyperpersonal model (Walther, 1996) suggests that such online deception, a form of selective self-presentation, may lead to misperceptions by potential mates. If the interaction continues offline, the receiver may feel disappointed or deceived by this manipulation.”

Skip it. Skip the whole thing. Whole journals, whole fields, produce nothing but this kind of stuff. All made possible by Statistics.

You’re welcome.

NEW PAPER: Why Models Run Hot: Results From An Irreducibly Simple Climate Model

This post is one that has been restored after the hacking. All original comments were lost.

Here’s our press release. See also the Daily Mail.

WHAT WENT WRONG?

A major peer-reviewed climate physics paper in the first issue of the prestigious Science Bulletin (formerly Chinese Science Bulletin), the Orient’s equivalent of Science or Nature, exposes elementary but serious errors in the billion-dollar general-circulation computer models relied on by the UN’s climate panel, the IPCC. The errors were the reason for concern about Man’s effect on the climate. Without the errors, there is no climate crisis.

The IPCC has long predicted that doubling the CO2 in the air might eventually warm the Earth by 3.3 oC. However, the new, simple model presented in the Science Bulletin predicts no more than 1 oC warming instead—and possibly much less. The model, developed over eight years, is so easy to use that a high-school math teacher or undergrad student can get credible results in minutes running it on a pocket scientific calculator.

The paper, “Why models run hot: results from an irreducibly simple climate model“, by Christopher Monckton of Brenchley, Willie Soon, David Legates and Matt Briggs, survived three rounds of tough peer review in which two of the reviewers had at first opposed the paper on the ground that it questioned the IPCC’s predictions.

When the paper’s four authors first tested the finished model’s global-warming predictions against those of the complex computer models and against observed real-world temperature change, their simple model was closer to the measured rate of global warming than all the predictions of the complex “general-circulation” models (see the picture which heads this post).

Next, the four researchers applied the model to studying why the official models concur in over-predicting global warming. In 1990, the UN’s climate panel predicted with “substantial confidence” that the world would warm at twice the rate that has been observed since.

ransom30

The very greatly exaggerated predictions (orange region of the picture above) of atmospheric global warming in the IPCC’s 1990 First Assessment Report, compared with the mean anomalies (dark blue) and trend (bright blue line) of three terrestrial and two satellite monthly global mean temperature datasets since 1990.

The measured, real-world rate of global warming over the past 25 years, equivalent to less than 1.4 oC per century, is about half the IPCC’s central prediction in 1990.

The new, simple climate model helps to expose the errors in the complex models the IPCC and governments rely upon. Those errors caused the over-predictions on which concern about Man’s influence on the climate was needlessly built.

Among the errors of the complex climate models that the simple model exposes are the following:

  • The assumption that “temperature feedbacks” would double or triple direct man-made greenhouse warming is the largest error made by the complex climate models. Feedbacks may well reduce warming, not amplify it.
  • The Bode system-gain equation models mutual amplification of feedbacks in electronic circuits, but, when complex models erroneously apply it to the climate on the IPCC’s false assumption of strongly net amplifying feedbacks, it greatly over-predicts global warming. It is the wrong equation.
  • Modellers have failed to cut their central estimate of global warming in line with a new, lower feedback estimate from the IPCC. They still predict 3.3 oC of warming per CO2 doubling, when on this ground alone they should only be predicting 2.2 oC—about half from direct warming and half from amplifying feedbacks.
  • Though the complex models say there is 0.6 oC man-made warming “in the pipeline” even if we stop emitting greenhouse gases, the simple model—confirmed by almost two decades without any significant global warming—shows there is no committed but unrealized man-made warming still to come.
  • There is no scientific justification for the IPCC’s extreme RCP 8.5 global warming scenario that predicts up to 12 oC global warming as a result of our industrial emissions of greenhouse gases.

Once errors like these are corrected, the most likely global warming in response to a doubling of CO2 concentration is not 3.3 oC but 1 oC or less. Even if all available fossil fuels were burned, less than 2.2 oC warming would result.

Lord Monckton, the paper’s lead author, created the new model on the basis of earlier reviewed research by him published in Physics and Society, in the UK Quarterly Economic Bulletin, in the Annual Proceedings of the World Federation of Scientists’ Seminars on Planetary Emergencies, and in Energy & Environment.

He said: “Our irreducibly simple climate model does not replace more complex models, but it does expose major errors and exaggerations in those models, such as the over-emphasis on positive or amplifying temperature feedbacks. For instance, take away the erroneous assumption that strongly net-positive feedback triples the rate of man-made global warming and the imagined climate crisis melts away.”

Dr Willie Soon, an eminent solar physicist at the Harvard-Smithsonian Center for Astrophysics, said: “Our work suggests that Man’s influence on climate may have been much overstated. The role of the Sun has been undervalued. Our model helps to present a more balanced view.”

Dr David Legates, Professor of Geography at the University of Delaware and formerly the State Climatologist, said: “This simple model is an invaluable teaching aid. Our paper is, in effect, the manual for the model, discussing appropriate values for the input parameters and demonstrating by examples how the model works.”

Dr Matt Briggs, “Statistician to the Stars!”, said: “A high-school student with a pocket scientific calculator can now use this remarkable model and obtain credible estimates of global warming simply and quickly, as well as acquiring a better understanding of how climate sensitivity is determined. As a statistician, I know the value of keeping things simple and the dangers in thinking that more complex models are necessarily better. Once people can understand how climate sensitivity is determined, they will realize how little evidence for alarm there is.”

Contact Christopher Monckton of Brenchley, monckton@mail.com, +44 7814 556423 or +44 131 225 5551.

——————————————————————————-

Well, press releases are press releases. Download the paper. Read that.

« Older posts Newer posts »

© 2016 William M. Briggs

Theme by Anders NorenUp ↑