Benoit Mandelbrot, RIP

Benoit Mandelbrot the great mathematician has passed away aged 85.
(HT: Marginal Revolution)

This entry was posted in Uncategorized. Bookmark the permalink.

42 Responses to Benoit Mandelbrot, RIP

  1. TerjeP says:

    I wrote my first Mandelbrot set program when I was about 14. It was written in basic on a Toshiba MSX128 and it took hours to render a basic image but I was hooked. I wrote a new version from scratch in uni using a combination of pascal and machine code (dividing by two can be executed faster by multiplying by 0.5 but there is nothing like a bit shift or exponent increment for speed). About a fortnight ago I rediscovered the Mandelbrot set via a free program for the iPhone. What enthralled me when I was 14 still enthralls me, which is the notion that a universe of infinite complexity and amazing beauty can be generated in the palm of your hand with such a rudimentary equation. It makes plausible the notion that all of existance is merely a rudimentary algorithm unfolding it’s hidden wonder.

    About the man behind the set I know a lot less.

  2. C.L. says:

    I confess that I don’t understand a single word of what Terge just said.

    Lots of brainy folks at this blog.

    🙁

  3. JC says:

    He’s just showing how geeks get off on smallish things, CL.

    We should be nice to them, as they do lots of really smart stuff we can then use to make our lives better.

    The world is a better place with lots of virgin geeks working hard on this sort of stuff.

    Let them be I says.

  4. Andrew Reynolds says:

    Terje,
    The Mandlebrot set I have always seen as interesting, but his work on financial markets is (IMHO) more interesting, even if it has unfortunately been substantially ignored. He was the first to show that a Gaussian normal distribution is inappropriate to describe the behaviour of markets and that a Levy distribution (with fatter tails) is a more appropriate fit to the data.
    The pity is that he has been ignored on a lot of this – importantly by the regulators, who have been using normal distributions in analysing risk.
    May he rest in peace – after a long and very productive life.

  5. . says:

    Shouldn’t ALL statistics ideally be non-parametric?

  6. Sinclair Davidson says:

    . – in many cases the difference between parametric and non-parametric doesn’t matter – you get the same result.

  7. Jc.. says:

    Andrew

    How can you use anything other then normal distribution in assessing risk. Anything else and it would be little more than mere opinion.

  8. Andrew Reynolds says:

    If you start with the right assumptions about the distribution, Sinclair. Mandlebrot showed that starting with “Assuming a Gaussian Normal distribution…” is the wrong way to start – and is based on late 19th Century work.
    I did a piece that resulted in a long thread at Ozrisk some time ago.

  9. TerjeP says:

    I must admit that I was still a virgin when I wrote that first Mandelbrot set program. However I suspect that a lot of less geeky 14 year olds were also.

    CL – if your desperate to understand take a look here:-

    http://en.m.wikipedia.org/wiki/Mandelbrot_set?wasRedirected=true

  10. Andrew Reynolds says:

    JC,
    We crossed. Have a look at the Ozrisk piece I linked to.
    The Normal distribution is not far out – it is plenty good enough while volatility is low. It does, though, understate the observed probability of extreme events. Mandelbrot showed this in the early 1960s.

  11. TerjeP says:

    Is Jc.. the same person as JC?

  12. C.L. says:

    The Mandelbrot set is a mathematical set of points in the complex plane, the boundary of which forms a fractal. The Mandelbrot set is the set of complex values of c for which the orbit of 0 under iteration of the complex quadratic polynomial zn+1 = zn2 + c remains bounded. That is, a complex number, c, is in the Mandelbrot set if, when starting with z0 = 0 and applying the iteration repeatedly, the absolute value of zn never exceeds a certain number (that number depends on c) however large n gets.

    Thanks, Terge.

    That clears it up for me. 🙂

  13. This is going to come across as snide, I know, but really: CL, it’s remarkable that you have such strong opinions on a science subject like climate change when you’ve never heard of the Mandelbrot set. (Not that it’s directly relevant – it isn’t as far as I know – but it just speaks of a certain lack of interest in science and maths, even at the popular level.)

  14. C.L. says:

    Steve, you once claimed that Nick Minchin was responsible for the summer of 2009-10. You also claim repeatedly that there is a link between scepticism about “climate change” and the prostate.

    You are to science what Sir Robert Helpmann was to oxy-welding.

    I’m intelligent enough to you that warmenism is a faith-based ideology and the last time I looked, nobody was suggetsing we increase the price of food and electricity, shut down coal mines, spend trillions of dollars or run for the hills because of the Mandelbrot set.

  15. daddy dave says:

    I’m with terje.

    the notion that a universe of infinite complexity and amazing beauty can be generated in the palm of your hand with such a rudimentary equation.

    the short of it is that Mandelbrot invented a simple rule that created a beautiful, complex structure.

    and yes, Jc is JC, except he’s writing from a different computer, I think.

  16. Andrew Reynolds says:

    Steve – I would suggest you and C.L. take it over to the open thread and leave this one for discussions of mathematics, rather than Minchin.

  17. C.L. says:

    More accurate and fair:

    “Steve – I would suggest you and C.L. take it over to the open thread and leave this one for discussions of mathematics, rather than Minchin.”

  18. stackja says:

    Chaos theory owes a lot to BM

  19. Reynolds – CL brings up this dishonest guff about Minchin all the time. It’s just part of his repetitious diversionary tactics. I’ve no intention of discussing it further.

  20. Andrew Reynolds says:

    Whatever, C.L. It remains my suggestion – and even, perhaps, request.

  21. . says:

    “it’s remarkable that you have such strong opinions on a science subject like climate change when you’ve never heard of the Mandelbrot set. (Not that it’s directly relevant”

    The Chewbacca defence. Nice one, you chucklehead.

  22. AB says:

    I never comment here but I thought C.L. you might be interested in The Mandelbrot Set “The Thumb Print of God”.

    Arthur C Clarke-Fractals-The Colors of Infinity 1-6

    HT. Bolt’s blog

  23. JC says:

    Steve… you really are a meathead. You have absolutely no idea what the hell you’re talking about.

    It’s been statisticians that have actually opened up a can of worms, as a result of the bullsitting going on from the top echelons of the global warming movement.

    Recall how our very own CISRO were using dishonest statistics to promote the absurd idea southern Australia would turn into a desert as a result of a 3 billion year drought brought on by global warming?

    Recall?

    The guy at Niche Modeling blog- a capable statistician- showed how this was an incorrect assumption and put those dishonest tax eaters back in their place.

    (this year by the way Victoria has more freaking water over it that the entire load held in the Pacific Ocean).

    It was another statistician at Climateaudit showed how the hockey stick was a complete made up crock of shit.

    You have a nerve to be suggesting what you are. Real nerve.

    If anything it’s been shown time and time again how climate scientists aren’t applying statistical techniques to analyze data.

  24. Sinclair Davidson says:

    Andrew – in the late 90s I went through a stage of calculating both parametric and non-paramentric stats in many of my papers and it never once made a significant difference to the results. (I accept that there are instances when it might but that is my answer to .’s question, “Shouldn’t ALL statistics ideally be non-parametric?”).

  25. Chris M says:

    Oh wow, I use fractal based mathematics in some of the design work I do…. it’s not just about pretty patterns it does have engineering uses. Enjoy looking at the fractal patterns seen in nature also, many are really beautiful.

  26. Chris M says:

    And regarding chaos theory I thought one of the outcomes was the proof that it was physically impossible to accurately predict the weather patterns much more than a week or so in advance. So much for these idiot ‘climate scientists’ with their 50 and 100 years predictions of doom, tell us exactly what will happen next month first.

  27. dover_beach says:

    Chris M, that’s because weather is chaotic but climate is deterministic. By some miracle, a process that is at short-time scales chaotic becomes deterministic on longer time-scales. Well, that is the conventional wisdom anyway.

  28. Peter Patton says:

    Chris M

    Enjoy looking at the fractal patterns seen in nature also, many are really beautiful.

    I had a Maths teacher who was a raving Christian. He was attracted to Pure Maths because it best showed the beauty of God’s creation. I’ll never forget the cracker lectures he gave on Fibonacci numbers in nature!

  29. PSC says:

    The pity is that he has been ignored on a lot of this – importantly by the regulators, who have been using normal distributions in analysing risk.

    Andrew – is this literally true, or is it hyperbole?

  30. TerjeP says:

    Chris M – the most experienced seaman can’t tell what size wave will break on the beach even just an hour in advance. And yet the tide can be extremely accurately predicted decades in advance. Waves and tides are different and weather and climate are also different.

  31. Michael Sutcliffe says:

    And yet the tide can be extremely accurately predicted decades in advance. Waves and tides are different and weather and climate are also different.

    I don’t think that’s accurate. We know why we can predict tides: specifically, the movement of planets and moons which we can predict accurately and have done for some time. There is nothing that specific in terms of climate. The factors like planetary orbits that we can know with a degree of certainty have a much smaller effect on climate than the 100% effect it has on tides. In fact, I’d say our understanding of weather is probably better than that of climate. If you see a low pressure front coming through, you can say with near certainty and some hours warning that it’s time to batten down the hatches.

  32. JC says:

    Interesting Turge. So you are suggesting that the climate can be predicted 100 years hence in the same way a tide can be predicted long term.

    Wow. I suppose that’s easily done with a 150 years of dubious data and tree rings.

  33. JC says:

    Why am i thinking this is another one of Turge’s bold EU comment starters?

  34. Michael Sutcliffe says:

    I just want to say I’ve met Terje once and he is a really nice bloke, not to mention a committed libertarian with whom I share essential values!

  35. C.L. says:

    CL brings up this dishonest guff about Minchin all the time. It’s just part of his repetitious diversionary tactics…

    Steve, if you’re denying that you argued that Nick Minchin would be held accountable by the people for the heat of the 2009-10 summer, you are a liar. It was back when you were a foot soldier for your old hero, Kevin Rudd, and all the talk was the salvational ETS and the Greatest Moral and Economic Challenge of Our Time.

    As for repetition and “diversionary tactics,” I think your ceaseless commentaries about the prostates of “climate change” sceptics, Lord Monckton’s eyeballs and Tony Abbott mark you as the veritable crack addict of diversion addiction.

    Whatever, C.L. It remains my suggestion – and even, perhaps, request.

    Not “whatever” at all, Andrew. Your intervention was dishonorable in the sense that it is clear to everyone who introduced the extraneous matter to which you object. Confine your chastisements to that person, would be my advice, lest you come across as pushing an agenda. That’s the first point. The second is that I don’t require your approval regarding the nature of my comments here at Catallaxy.

  36. C.L. says:

    Thanks, AB. Your link is down but I found it at YouTube.

  37. Andrew Reynolds says:

    PSC,
    No hyperbole there. If you look at the Basel II Accords, for example, the regulators mandate the use of the normal distribution in the determination of regulatory capital – even helpfully giving the formula in terms of the relevant excel function. It is a pity they did not read even something reasonably populist like Taleb’s “Black Swans”.
    .
    Peter,
    You do not have to be a “raving” Christian to see the beauty in numbers. While I am, at best, indifferent (OK pretty useless) in higher mathematics I like to think I can appreciate the true beauty in a good proof or an elegant algorithm.
    .
    Michael,
    There is more uncertainty at the micro level that at the macro in many fields. Weather / climate is one, quantum uncertainty is another and, probably less so, economics is a third. The lack of Newtonian precision in any of these does not change that. Mandelbrot’s work (to the extent that I can understand it) showed that pretty well. The patterns in the chaos is the real beauty of that.
    .
    C.L.,
    Which is why I phrased my request the way I did. You will note, I hope, that I made my request after you responded to “steve”, which is why I included you both in it. Taking it to the open thread is still my request.

  38. TerjeP says:

    Michael – I don’t think climate is anywhere near as predictable as the tide. That isn’t what I suggested. What I was saying was that the fact that we can’t accurately forecast the weather does not stack up as a criticism of climate modeling. A better criticism of climate models is that to date none has made a notable prediction that has subsequently come to pass and validated the predictive power of the model in any significant way.

    I think my statement as follows was accurate:-

    Waves and tides are different and weather and climate are also different.

  39. C.L. says:

    Andrew, Steve tried to derail the thread – not me. You should have confined yourself to lecturing him. It looks like you were simply looking for an excuse to bait me. If you wish to do that, I suggest you take your agenda to the Open Thread.

  40. JC says:

    You left thus part out turge.

    And yet the tide can be extremely accurately predicted decades in advance

    which normal human being would read as forming part of your argument about climate prediction.

    But nice slice and cover. Next time just don’t get caught moving the ball.

  41. Peter Patton says:

    AR

    I did not say you have to be a raving Christian, I was merely providing narrative color and shade.

  42. TerjeP says:

    JC – the tide can be accurately predicted decades in advance. Waves can’t be. Either you have a basic problem with comprehension or else you wish to challenge this assertion. Which is it? Put up or shut up, please, because frankly I find you’re endless snarking tiresome. Nowhere did I say climate can be predicted as accurately as tides. Nowhere. Such an assertion would be daft. I merely asserted that a lack of ability to predict the weather does not imply an inability to predict climate. Climate and weather are different just as tides and waves are different.

    Why, why, why do you carry on as if you know what people “really mean” better than they do. Normal people ask a question if they are in any doubt. You just make stuff up and launch into strawman arguments and personal abuse. It’s quite astonishing because you are not unintelligent. Did you have a difficult childhood that left you feeling like the whole world is against you? You carry on like some overly hormonal teenager that thinks everybody that makes eye contact is picking a fight. Stop it.

Comments are closed.