Sunday, September 27, 2009

My Trading Philosophy

…are you saying that the markets are not likely to see any gross movement 24 hours after either an attack by Israel or Iran, or say, any movement, over the next 7 days, 1 month, 3 months, that it'd all happen in the 1st hour or so?

Clearly you are in need of 'boning up' on the "Efficient Market Hypothesis" (AKA, EMH)! There is a "hard" form and a "soft" form.

The hard form says that all public information/knowledge is immediately factored into market prices, so that whatever you do, once you take the risks into account, you are no better off than you were before such knowledge was made public! Despite its non-intuitive nature, this form actually was able to survive for many years (before it was proved to be technically) false. What that means, is that for all intents and purposes, you would need to be a Wall Street 'quant' in order to "take advantage of" any 'breaking' news.

This is more or less easily refuted (nowadays), since using a 'reductio ad absurdum' you can prove that the hard EMH will predict that since the market is 'always at or near equilibrium', little or no trading takes place (especially as all of that news will most probably cancel itself out)! Or that such trading as does take place is correlated to the amount of news breaking at any given time (but how does one measure that?) and is entirely random as to net market effects. It also assumes the 'absence' (improbability) of any large anomalies, so that it is constitutionally unable to come up with a reasonable explanation for something like the October 19, 1987 25% stock market crash! (The EMH explanation is that such an event is perfectly predictable as a very, very low probability event way out on the tail of the gaussian distribution, ie, a "once in the lifetime of the universe event"! Unfortunately, such events seem to be happening rather more frequently (sometimes within hours of each other— much to the surprise and chagrin of the 'quants'.)

The soft form allows that stock market prices may not immediately reflect the news, but on the average over the long run, you would do no better than chance (ie, your net gain would be zero) by trying to beat the effects of the news (sometimes the effects will go one way, sometimes the other)! A terrific example of this is what happened to the Wall Street 'quants' in summer of 2007. For almost 5 years, the 'quants' had seemingly beaten the averages by using incredibly obstruse mathematical algorithms (which no one on Wall street understood except them), super-computers, "flash trading", and similar techiques to take advantage of very, very minor "anomalies" in stock market prices that appeared only briefly (before traders such as the quants erased them). Then in the space of days, they lost all of the money they had gained— and then some— when the market 'perversely' turned against their models, causing the stocks they had shorted to go up and the stocks they held long to go down! A double whammy which was due to the fact that their models had assumed no correlation among prices but, in fact, the very (similar) models the quants created produced exactly such a correlation! Moral: the market is not infinite, is not gaussian, and is prone to (correlated) panic! (P.S. The crashes of 2008 and March, 2009 improved their education still further!)

I do not hold with EMH (insofar as I have any overriding philosophy of the market, I am a behavioral finance theorist); however, I have never been able to refute the 'soft' version of EMH (nor, to my knowledge has anyone else). So, I adhere to the principle that long term investing and avoiding "running with the pack", aka, "MOMO" investing, is best. That is, I use a "contrary", 'informed' investing approach that will produce positive results in the long run, because most investors are quite short term oriented and heavily influenced by what others are doing! (So, I am taking advantage of an "anomaly" which most other traders are not willing to exploit.)

However, I very strongly do not believe in the 'Buy and Hold' investing approach, much favored by Wall Street for 'naive' investors, especially as I very strongly believe that the market is cyclical in nature— though as with all things involving people, hardly perfectly so. As various people can attest, I have been predicting something close to financial Armageddon from 2001 on (though I was a little late; I expected it to occur around 2008/9… and not 2007/8…). This was largely because of the state of the financial markets (it was clear already in 2001 for anyone willing to see that we were absolutely in "melt up mode" financially that made the twenties look tame by comparison and could only have ended the same way!) and because the market has a very strong tendency to "echo" itself every forty years or so (2009 is 80 years after 1929 and the horrible markets of the '30s, and 40 years after 1969 and the horrible markets of the '70s)! Indeed, the horrendous bear market of 2000-2002 merely echoed the bear market of 1920-21. And, yes, 1960 was a (mild) recession year.

For my serious money, I prefer to use Sy Harding's STS timing approach… Check it out. I have been aware of the 'seasonal effect' in the stock market for over 40 years; but that effect was too marginal to use for trading, until Sy added his 'tweak'.

No comments:

Post a Comment