Feb

24

 I take the view that most market predictions don't produce statistically sound results.

I believe market condition is a result of human behavior. Although many fundamental human behaviors are predictable with the advancement of behavioral sciences, the market involves more than the fundamental human behaviors. The key to it lies in the varying derivative perceptions of market participants.

Let's say the view "since condition A, then the market will rise" is the fundamental perception. A first derivative view, for instance, can be "since all believe the market rises due to condition A, it will fall". A second derivative can then be "since all believe the market falls due to (…), it will rise". And so on.

If the market participants all adhered to one of the principles above, then it would be very easy to predict the market. The thing is that is never the case. The big hands shift views along the derivatives all the time. That is what makes most predictions today unsound.

If we have a way (the big winners should have this talent) to predict which derivative view the big hands take, then the prediction would be more accurate. But then, if many could do that, the game changes again.

Before that happens, let me raise the question here on how one can develop that talent.

Steve Ellison writes: 

This is the principle of ever-changing cycles, as described by Bacon in Secrets of Professional Turf Betting and elaborated on in the Chair's books.

One of Bacon's approaches was to look for good horses that had lost in their most recent races. The memories of the recent losses caused the public to have negative opinions about those horses.

George Parkanyi writes: 

Market movements (which way, how far, and when) cannot be predicted by DEFINITION. The financial markets are a non-linear system. More than three non-correlated variables, and you cannot predict a specific price at a specific future point in time — a mathematical certainty.

However, like many natural cycles, markets exhibit a powerful tendency to revert to the mean. Now that mean moves around and will have its own wobble, but around it there is definitely a clear sinusoidal pattern — actually short and longer term patterns within patterns. Look at any index or commodity chart over an extended period of time. Individual securities will have the same tendency, but the longer term impact of low-probability outliers is more pronounced — your Microsofts or your Lehman Brothers'. The more narrow the influences — the greater the risk (one-trick pony, or bad management risk for example.) If you apply portfolio theory (diversification basically), the risk (and reward) of outliers is significantly diminished, and I believe you can develop successful strategies simply based on price and on the concept of reversion to the mean using indices, sectors, and commodities (anything always economically necessary to greater or lesser degree, that has an extremely low probability of going to zero. Even there it's not quite blow-up risk-free (asbestos didn't really work out.)

You can then more safely use leverage and modulate the range of returns with same. Reversion to the mean requires a large sample size, so you need many sources acting on the main influences on price (large underlying product markets, liquid financial markets), and time. The larger the sample size and the longer the time frame, the more reliable a reversion strategy should be.

The psychology of what Steve alluded above to is reflected in the aggregate behavior (influences on price) mentioned above. If you're going to go for the "bad horses" though, buy several in case one keels over and dies. Now if they all contract a contagious disease from each other…

George Coyle writes:

The 86th episode of Seinfeld was called "The Opposite" and involved George Costanza's experience in cycles.

The plot goes as follows (from wikipedia):

George returns from the beach and decides that every decision that he has ever made has been wrong, and that his life is the exact opposite of what it should be. George tells this to Jerry in Monk's Cafe, who convinces him that "if every instinct you have is wrong, then the opposite would have to be right". George then resolves to start doing the complete opposite of what he would do normally."Of course everything goes right for him from then on (until the end of the episode). While funny it brings up the interesting idea of the Costanza trade. Out sample seldom replicate in sample (probably due to ever changing cycles). How can one figure when to follow the trend of profitability and when to apply the Costanza trade to a perceived winner. 


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search