Thursday, April 23, 2009

Nassim Nicholas Taleb david hume and your portfolio

I suppose it's my fault for just listening to the lecture.
but i looked at the bio for Prof. Taleb, and it's quite fascinating.
With an MBA, he worked as a floor trader and senior trader at some of the biggest trading houses on Wall Street. During the course of his career, he earned a Phd. from the university of Paris.
What Taleb was concerned with was no esoteric formulas about the market, or how to hedge properly (though he did write a book called Dynamic Hedging). taleb is interested in one thing and one thing only: chance.

Not probability of an event, but the impact of the event will have on probability. He is much more capable of describing chance than I am. But I visualize it in this way:

For an investor, or trader or anyone in business or anyone trying to predict the future in any shape or form.

When you try to make a bet about the future, when you try to make a prediction based on passed observations, you are using a theory.

theory is very limited in it's use. why? because of chance. because of the fact that when it comes to social, cultural, political events, natural disasters, wars- we simply cannot predict when these things will happen.

these events that take place in the world really do have consequences for society. The market meltdown we have all just witnessed is the perfect example.

the vast majority of people put their faith in investment advisors, mutual funds and money managers, who had in turn put their faith in financial institutions that hand in turn put their faith in excessive borrowing and putting borrowed money into bad investments.

The danger is that when a cataclysmic event, like the credit crisis, hits, human beings are genetically and biosocially predisposed to reconstructing our original predictions and forecasts in the presence of new observations.
do people actually do this? well there is lots of clinical data to support the fact that when an individual or groups of individuals make predictions, observe the data, and then later reflect abou their predictions. it doesn't matter how wrong their predictions were from the observation. Clinical studies show people reconstruct their past beliefs to fit with the current reality. I know I'm guilty of shaping how I see many of my past experiences differently at different points in my life.

the point of saying all this highlights the fact that economic modelers, economic forecasters, people who take data from a year or the past century or whatever, and then try to use excel or some other fancy mathematics to extrapolate the past data into a model of what to expect are NO BETTER THAN ASTROLOGERS. This goes for the majority of investment advisors, money managers and do-it-yourself amateur investors. All of these people who use increasingly elaborate mathematical economic theory to predict what will happen in the market place lost a great deal of money over the past 6 months. So has every single person who followed the 'belief system' of the world that existed before october of 2008.

Before october 08 every single investor believed that equities were always on a one way ride up. There could never be a credit freeze again (we were too smart, regulators were too well-informed) and we just kept getting better at everything, winning in Iraq.

It made it easy to deal with paying 3x the amount for oil and natural gas than the year previous.
people didn't think about the long-term consequences of a massive and rapid run-up of energy prices. They didn't think about it because the last time it happened the stock market wasn't wiped out by half.

The Black Swan is not just a book, it is an ancient philosophical problem. Put in contemporary terms, it was a common phrase like 'when pigs fly' to say that something was as likely as a black swan. Well, this phrase was used before the British had discovered Australia, where they did indeed find black swans. David Hume called his the problem of induction. when we create a theory or law, based on repeated observations. The essential question proposed by Hume was: what is the real law or guarantee that just because i have observed the sun rising and setting every day of my life and for those who have ever lived they say the same thing. What hard and fast law says that it must rise tomorrow or that it must always rise every day? From a strictly philosophical point of view, he is correct. the problem of induction is one that philosophers have never answered and never posed seriously because it essentially undermines the entire academic world. the problem of induction says that we can essentially predict nothing. we can have no absolute rules for our world in any domain and therefore the world as we observe it is completely unknowable. it's not even just unpredictable or chaotic, it's unknowable.

Now, that logical conclusion is what led many great philosophers, who tried to tackle the problem of induction head-on, to become deeply religious men.

personally, i take a bit of a positivist attitude. There are real observations that we can make and there are limited theories and predictable measures of behavior across a large number of domains.

Taleb refers to two domains of knowledge and existence: mediocristan and extremistan.
mediocristan is all the stuff we learn in schools but is of no practical value
extremistan is like how the real world works, where only a handful of people control most of the wealth in the world and evey body else has only a smaller average.

mediocristan is the domain where if you get enough numbers and do enough calculations you should be able to do accurate models and predictions. This is the domain of dentists, where even the best dentist in the world, his income doesn't differ that wildly from the average dentist income.

extremistan is where warren buffet and george soros and a few other guys control half of Wall street and hundreds of billions of dollars and the millions of the rest of the investment world is trading in a few thousand bucks here and there.

the point about the difference between mediocristan and extremistan: we like to believe the world is like mediocristan (and in many aspects of our life it is that way) but we generally ignore and limit our understanding of the domain of extremistan.
we prefer, through listening to anyalysts and forecasters and gurus, to ignore the fact that extremistan is a real domain of life and it has very harsh realities. the realities that only a few dominate the average and that it is totally unpredictable as to who will be the dominant in extremeistan.
there is no financial model to predit the success and value of a google.
there was no financial model (outside of a few astute observers) for the recent financial meltdown, untill after the fact.
further evidence to the idea that extremistan (which is primarily the domain of economics and finace) is totally unpredictable, for even the most dominant, is Warren Buffet. he spent a princely sum of money on a giant oil company just before the giant crash in oil. maybe he did this on purpose, but that's unlikely in any scenario.

the lesson for your investments: if you think stocks and trading in other forms of equity-based investments are a safe place for the majority (or even bigger than 10%) of your savings, if you have been given this advice from an advisor or read it from an investment guru, it is the same as if you had taken the advice from an astrologer.
the data, as discussed by Taleb, shows that when it comes to economic forecasts and predicting what the market will do, human beings simply fail time and time again. Life and finance are too complicated to ever be able to know if or when the financial black swan will arrive.

therefore, look at my previous post and start to think about what's really in your best interest.
extremistan and the fact that i can't know anything about the future and that's the one thing i can know about it, makes one prone to the ultra-conservative with investments. prepare for the worst in all circumstances and take some money and go nuts with it.

1 comment:

  1. by biosocial i mean that there is a feedback between what we feel and do in our body and habitat and how we react with the community or wider society. difficult concept to explain, but i would just point out that it's not the social darwnism type of theory that says human biology determines the organization and success or failure of a society. That is freudian hogwash and has been discredited since the Nazi's.