Mca 6 2007 def pdf
Download 192.28 Kb. Pdf ko'rish
|
MCA200706121
Professor Taleb, how would you introduce yourself to our readers? I consider myself a scholar, who specializes in un- certainty. I used to be a stock trader for a very long time, but since the crash of 1987 I luckily did not have to worry much about the more mundane. So I specialized in complex derivatives on the one hand, and probability theory on the other. I looked at the errors in judgment when dealing with probability. I have a PhD in applied statistics, but from ’87 this interest evolved into an interest in the role of large events. But: seen from all kinds of perspecti- MCA: oktober 2007, nummer 6 10 Interview: NASSIM TALEB: YOU CANNOT BECOME FAT IN ONE DAY, BUT YOU CAN BECOME ENORMOUSLY WEALTHY! On November 7 the Controllers Institute will hold it’s annual conference. One of the more interesting, but maybe less well-known keynote speakers will be Nassim Nicholas Taleb , author of bestsellers Fooled By Randomness and more recently The Black Swan . Taleb divides our environment in Mediocristan and Extremistan ; two domains in which the ability to understand the past and predict the future is radically different. In Mediocristan, events are generated by an underlying random process with a ‘normal’ distribution. These events are often physical and observable and they tend to cluster in a Gaussian distribution pattern, around the middle. Most people are near the average height and no adult is more than nine feet tall. But in Extremistan, the right-hand tail of events is thick and long and the outlier , the wildly unlikely event (with either enormously positive or enormously negative consequences) is more common than our general experience would indicate. Bill Gates is more than a little wealthier than the average. 9/11 was worse than just a typical bad day in New York City. Taleb states that too often in dealing with events of Extremistan we use our Mediocristan intuitions. We turn a blind eye to the unexpected. To risks, but also, all too often, to serendipitous opportunities. Editor Jan Bots talks with Taleb about chance, risk and how to cope with it in the real world. MCA: oktober 2007, nummer 6 11 ‘A Black Swan is an event with a very small probability, but huge repercussions’ ves – philosophy, psychology, economics, history, social sciences, all blended together. I was never interested in finance per se. I was interested in what I could learn from finance. When people ask you ‘What are you?’ they usually mean ‘What job are you being paid for?’ Now no- body pays me a salary… I do what I do. Not every reader will be familiar with your two books, Fooled by Randomness and The Black Swan. Could you summarize their message for us? Really, the Black Swan summarizes it all I think. Rare events, what I call Black Swans, play a massive role. A Black Swan is an event with a very small pro- bability, but huge repercussions. Before it happens we never even think about it. Then, after the fact, we realize it really had been predictable, but that our science was not very effective: we apparently have a huge amount of retrospective predictability skills, but no effective prospective ones. These rare events play an monumental role in our lives – whether in finance, in economics, poli- tics, war and peace, or inventions. The world as it appears is not very predictable, because it depends on concentrated random variables. We may think we can model it; we cannot, because the outliers do- minate. In history books you’ll read that the first world war was predictable. But if you look at war bonds, it was not. They predicted 1870, but not 1914. This war was only retrospectively predictable. People were lulled by a century of post-Napoleonic peace. One metaphor I use to explain this is book sales. In 1995 there were maybe 10.000 books published in the States. 70% of the sales revenue was generated by five titles: outliers, statistically seen. Likewise, Bill Gates is infinitely more wealthy than most other people. If you consider the role of large deviations in wealth, market share, casualties in war, it is so potent; it makes everything unpredictable. And how do we cope with this unpredictability? We can do a lot about this, if we start to perform what I call domain separation. There are two kinds of domains. One, in which a large deviation plays an inconsequential role, what I call mild random- ness. In the other domain, of wild randomness, de- viations play a massive role. Eating is an example of mild randomness: the quantity of food you can ab- sorb in one day is limited. You can not become fat in one day. This is the world of predictability. Becom- ing rich or poor is the domain of wild randomness. You can easily become rich (97% of the money I ever made came in one day, in October 1987) or poor (war, earthquake) in a single day. How do you know or determine in which domain you are? Pretty much everything socio-economic is in the domain of wild randomness, especially when it is linked with information. There are ways to do diag- nostics from the role of random variables. For in- stance with books: you just calculate the share of the minority in the total. Or you can perform an ex- ante analysis by asking the question whether there are natural limits to the variable, as in the height or weight of people. In socio-economics you seldom have natural limits. But if I take the example of book publishing, my publisher knows his rule-of-thumb: that only 2 out of 100 titles will be a success. Yes, but he does not know beforehand which ones. Nor will he know whether he’ll sell a thousand or a million copies. In some years 70% comes from seven books, in other years 50% from twenty. You can pu- blish books for ten years and never have a hit. Fact is, our ability to generalize from sample sets is monstrously low. If you look at biotech, can- cer research for example, we have been doing that for 40 years now without any real breakthrough. The last one, chemotherapy, by the way was an se- rendipitous invention. How long did we have to wait for Microsoft? Another example: Google. We are living in a world in which first of all there is great uncertainty about these random vari- ables. What I noticed in my research, and that is more or less my business now, is the following: I probably cannot figure out all these probabilities, but what we are very good at is determining whether my world is sensitive to these random variables, to an outlier or not. That is very easy to ascertain. That then is the big question? Yes. Take portfolio’s. There are portfolio’s, sensitive to what I call the negative Black Swans, and portfo- lio’s that are sensitive to positive Black Swans. MCA: oktober 2007, nummer 6 12 ‘There are ways to check the robustness of portfolio’s’ MCA: oktober 2007, nummer 6 13 Did you know, in 1982 banks in America lost cumu- latively 800% of all the profits made in the history of banking. Of course with the consent of Paul Vol- cker this was kept hidden. The eight biggest banks in America had $ 22 bn in capital, and $ 60 bn in loans to South and Central America. They lost everything in one incident. You see, you can tell right away if you are in an environment that is sen- sitive to negative Black Swans. Ten years later, with the Savings &Loans fiasco, they made the same mistake again! This one cost us 700 bn dollars. And now they may do it again. In other words, banking is an industry, very sensitive to Black Swans. Now let’s take an industry that is sensitive to positive Black Swans, real estate. Real estate has been, in America and elsewhere, very profitable, Download 192.28 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling