It occurs to me that I still haven't scored and posted the winners of the BoF 2008 Election Prediction Contest. Formally I still can't, because the Minnesota senate race has not legally been resolved. For all practical purposes it has, but I like to be sure about things before I go and award the contest to someone.
But while we're waiting, how about another contest? Here it is:
Predict the lowest close of the Dow Jones Industrial Average occurring on or before January 1, 2010. Trading during the day doesn't count, and it doesn't matter on which day the minimum occurs. Just the value at the close.
I would like to think this minimum will serve as an infimum for all subsequent trading days, but there is such a thing as being irrationally exuberant.
Predicting the stock market is voodoo of a sort physicists occasionally like to BS about. There are quantum events which are entirely random so far as we can tell, and there are perfectly deterministic classical events (such as positions in the three-body problem) which are so computationally difficult as to be random for all practical purposes. The stock market is not all that closely analogous to either, but it is as least very unpredictable.
The efficient market hypothesis is something I strongly suspect is true. It doesn't say that the market accurately prices stocks based on future value, or even that the market's participants are rational. It does predict that any known way to predict the market will result in people pricing those predictions into the market themselves. As such, it's impossible to outperform the stock market except by luck (or cheating, as in insider trading). Experience bears this out: professional investors and mutual fund managers almost never outperform the underlying market over the long term. If they can't, you can't either. The market is just too random. Staple the stock pages of your newspaper to a dartboard and have a monkey throw darts at it. You'll probably do as well as your mutual fund - better even, since you won't have additional fund loads. Seriously, the experiment has been done many times.
Of course there seems to be the occasional exception around the edges when new innovations like algorithmic trading begin to become common. But these are rare and generally transient phenomena.
While it's never popular to recommend a book that says "Sorry, you can't beat the system.", I'd say you owe it to your fiscal sanity to read A Random Walk Down Wall Street
- Log in to post comments
It's going to be a race to see who can be most pessimistic.
I'll chip in at 4501.
5,832 = 18^3. Seems about as good as any other number.
I predict it will be 6100.
My prediction: 5500.
I have no logic built into that number, it's completely a wild assed guess.
5500.
Uncle Al submits 3400. Prior posters omitted urban riots after social programs collapse. War is the traditional counter to economic collapse - don't outsource! Remove cities' inner ulcers then rebuild to recreate the economy and its employments.
I will guesstimate a lowest Dow close this year of around 4500. But I think it is overly optimistic to say that that will be the overall bottom. So far we have been repeating the mistakes of Japan's "lost decade". I would not be surprised if the Dow remained in the 4000-5000 range for several years, and it could go even lower.
Regarding the efficient market hypothesis, this is a reasonable guess for an academic, but there are practical people who violate it on a day to day basis. This is how they earn their living.
There are many things in life which we believe only because we have immediate evidence in favor of it. If I told you that it was possible to put a billion transistors onto a piece of silicon and then get it to play chess better than any human on the planet you would call me a raving madman unless you happen to know that these things have been done. The same analysis applies to those obscure talents that most people are unaware of. Intellectuals tend to confuse the fact that they do not know how to do something with the much more difficult to prove speculation that no one knows. Avoid this.
Learning to trade the market is easier than grad school but not by a hell of a lot. It's not something you can do overnight. Best is to use books (few are any use), teachers (only those who do this for a living please), and practice. I suggest budgeting for 2000 transactions over a period of time of about a year. That should be enough to learn how to trade, and importantly, also to gather sufficient statistics that you will be able to distinguish between the situations which are profitable for you from the ones that are not.
People can do this without learning mathematics, but you might try learning elementary statistics, and looking up "gambler's ruin" and "money management" in terms of how one chooses the size of a bet.
Thousands of people are paid to trade by firms, many thousands more trade on their own. A trading firm will teach you the rules but for a physicist you will probably understand it better if you learn the theory in a theoretical way. This is why Wall Street hires people with physics degrees.
The first step is getting accustomed to the market. Plan to spend a couple months doing nothing but watching the numbers go by all day long. If you have a real trading job, you will be assigned to watch the successful traders trade all day. This is not as boring as one might imagine. You will have a screen which shows their transactions.
At first, the activity in the market will appear completely random. You will have your efficient market hypothesis confirmed for you, except for the bizarre fact that the traders you are watching are violating it. Gradually, you will realize that even though they are trading different stocks, they all tend to buy and sell at the same time. You will have no clue why this is so. It will appear miraculous.
If you spend a few days in the forest, you will see the creatures of the forest do random things. You will not see any patterns or understand what is going on. As you spend more and more time there, you will gradually begin to understand cause and effect. Eventually you will be able to turn to your companion and say "see that tree over there? In a few minutes a flock of birds will land there." Your companion will think that you are in violation of the efficient ecology hypothesis and conclude that you were just lucky but you will know.
The markets have a sort of ecology. The energy source is "order flow". This consists of money coming into the market from all those drones that have automatic contributions to 401k plans, the insurance companies, etc., etc., etc. Different ecological niches are filled by people who have different time horizons. Generally speaking larger amounts of money must be traded on longer time horizons.
The thing one must constantly fight against is the efficient market, which tends to move prices randomly, which means proportional to the square root of time. One last word, always remember that to make money in the market all you have to do is pick out rare situations. Since the market is so large, rare situations occur frequently enough.
Remember that you're competing with other humans and also machines. But the truly excellent end up managing large amounts of money. This means that they cannot compete in the ecological niches associated with short time horizons.
The major downside to this sort of career is that it is incredibly, almost unbelievably boring. It is like hunting, you must spend most of your time watching and waiting. If you are looking for excitement then you are a fool and you will lose your money.
Good hunting.
Interesting post Carl, but you forgot to give us a prediction, O.K. maybe guess. You seem to have some background in trading and I would find your number interesting. So far I am the most optimistic in my prediction, but the later posters are giving me pause.
In any case thanks for your comment.
4847, and though it's not required, September 17. From that low it will quickly rebound and on January 1, 2010 it will be at just under double this year's low.
I'd probably gouge my eyes out by the time I reach 2000 transactions a year. And there would be no federal govt to bail me out.
If I post my prediction publicly, it will not come true. You'll just have to believe me that I was right.
Dow 950. Social unrest, societal collapse (making being the winner of this poll utterly useless and trivial). Third world status, martial law declared. Empty grocery store shelves and ensuing food riots (because Americans can no longer provide for themselves by sustinence farming). Americans rounded up into FEMA camps to keep them from killing the banksters who engineered the crisis.
My working assumption, looking at a long term graph of the DJIA is that, without the financial and real estate bubble, we would be somewhere between 6000 and 6500. That the fundamentals are something like supported in this range and that as teh average crosses this range the losses on the DJIA will slow, overshoot, and then return to the 6000 to 6500 range.
So I guesstimate 5800.
Anecdote:
A very long time ago, at a camp of some sort, I took part in a quiz game involving two teams of players. It was our team's turn to answer a question, and the question was: "What is the Dow Jones index?". One of us suggested a fanciful hypothesis along the lines of "price of eggs in China divided by price of real estate in New York", or something even more convoluted and unlikely, because we didn't have the faintest idea.
Naturally, the question went to the opposing team, who simply said, "It's to do with the stock market, isn't it?" and were awarded the point. Our cries of "We could have told you that!" failed to influence the quizmaster's decision.
I'm sure there's a moral in this story.
4357
The markets match the information fed into them. The person doing that initial feed can com out ahead. For instance: if a startup has a genuine useful invention that will make sqilions. Or: a new source of ore is discovered.
But tracking that sort of thing takes effort. Efficient market means that you cannot make money of the market without feeding new information into it about the real world. And that's what investors are suppoed to be doing; not watching wiggly lines generated by the market itself.
Interesting thought! Monkey business.
Herbert Simon did a lot to promote his position that the classical idea of 'rational man' doesn't work in economics. Dan Kahneman (and the late Amos Tversky) did even more to point out in what ways people make choices and decisions given that they are 'bounded rationalists'. Both Simon and Kahneman got a Nobel Prize for Economics. Presently further work on bounded rationality, specifically in the domain of 'frugal heuristics' or rules of thumb that work as good as or better than econometric models, is done by German investigators, predominantly by psychologist Gerd Gigerenzer and economist (also of Nobel fame) Reinhard Selten. In fact Gigerenzer showed that (and why) chimps perform better in stock picking than established marketeers and analysts.
A good starting point for getting acquainted with Gigerenzer's work is:
Gigerenzer, G., and Selten, R., (Eds.). (2001). Bounded rationality: The adaptive toolbox. Cambridge, MA: MIT Press.
Here's a pseudo-paradox about efficient markets: The market is only efficient if there are lots of people who are trying hard to exploit its inefficiencies. If everyone just threw darts to pick their stocks, then the stock market wouldn't be efficient. And if it's not efficient, there is room for someone to make lots of money by doing something smarter than throwing darts.
DJIA minimum on or before 01012010 6900. Why so optimistic vs this board? Well quantitative easing and printing money is going to cause massive hyperinflation which is going to kill cash positions forcing money into equities to perserve its value. Strong companies producing crucial needs will be more valuable than depreciating paper so their value will stay up when priced in units of that paper... ie inflation increases the DJIA as well as the cost of bread.
In today's dollars that number would be more like 5000