Local Predictability
by theorangedog on Jan.16, 2008, under Skills
Inductive reasoning in trading may be the preferred method for developing systems. One reason, as argued by Doyne Farmer in Cracking Wall Street, is due to Local Predictability. The concept is based in chaos theory, and references the idea that there may be an underlying order in the short term that enables prediction of events that are not equally predictable, if at all, in the long term.
Farmer was a founder of Prediction Company, which was purchased by UBS. There is an interesting presentation Prediction Company created a few years back, titled The Business of Model Based Trading. Both the article and the attached .pdf are worth a quick read.
In the referenced article, there are a number of analogies and layman interpretations of the interaction between physics, math, and finance. Below are two passages that I enjoyed:
He likes to use a favorite example when explaining the anatomy of a prediction. “Here, catch this!” he says, tossing you a ball. You grab it. “You know how you caught that?” he asks. “By prediction.” Farmer contends you have a model in your head of how baseballs fly. You could predict the trajectory of a high fly using Newton’s classic equation f=ma, but your brain doesn’t stock up on elementary physics equations. Rather, it builds a model directly from experiential data. A baseball player watches a thousand baseballs come off a bat, and a thousand times lifts his gloved hand, and a thousand times adjusts his guess with his mitt. Without his knowing how, his brain gradually compiles a model of where the ball lands — a model almost as good as f=ma, but not as generalized. It’s based entirely on a collection of hand-eye data from past catches.
And:
While running from lions, or investing in stocks, the tiniest edge over raw luck is significant.
Leave a Reply
-
Welcome to theorangedog.net.
This site is my dashboard for links and information on topics related to Skills (Finance), Spirits (Whisky & Beer), and Speed (Races & Training).
-
.
Check these guys out!
January 19th, 2008 on 7:12 am
Nice new site here, I like it.
I especially liked the last quote, “While running from lions, or investing in stocks, the tiniest edge over raw luck is significant.”
However, not many people know how to transform that tiniest edge into profitability. That’s the rub.
January 19th, 2008 on 12:23 pm
Great point Tom. My argument is that profiting from an edge is also a function of position sizing and risk management (alternatively money management), and in many cases those factors can improve one’s edge.
January 19th, 2008 on 1:36 pm
Statistical Arbitrage is one area where the idea of an “edge” takes a different form… some research has shown that the the returns distribution is quite positively skewed with sometimes non existent fat tails… particularly forex. Don’t know about stocks, but Higher frequency data is where most of the edge lies in modern day finance don’t you think?
January 19th, 2008 on 1:53 pm
It’s not something I’ve formally looked into, but I would say that there is a distinct dichotomy in the both the arena and detection of exploitable edges.
The first is what you mention, aiQUANT. High-frequency time series enables the detection of patterns, and I think there is currently an arms race for speed, and that will determine, at any given instance, who has the edge. Groups like RenTec and BGI come to mind (although BGI, to my knowledge, also still does a lot with fundamentals). Wall Street and Tech (see Web Resources section) has had numerous articles about the chase for speed, including some interesting ones that discuss I-banks and funds leasing space as close to the exchanges as possible for server storage.
The second arena, I would argue, is still in longer interval data, but with the edge being developed through behavioral finance. For example, look at many market corrections that have led to profitable trades because they were over-corrections. Equity issues such as MRK, AMR, MSO, MSFT, etc. come to mind, as well as some commodities such as CL (oil). Each of these has led to profitable trades for me, personally, and I think that the opportunities based upon behavioral finance still exist.
I’m sure what is obviously apparent, especially based upon my recent posts, is the split between methods in identifying the edge in these arenas. High-frequency requires, I think, a highly inductive process, while low-frequency requires a deductive method.
January 24th, 2008 on 11:55 am
Is HF trading just market-making on crack? I’m being (halfway) serious …
I believe the longer-timeframe behavioral finance edges still exist and will continue to persist because the markets don’t change that much, and people (as a class, not as individuals) don’t change at all.
Inductive and deductive reasoning only exist in pure form inside of textbooks. The process is inductive-deductive-inductive-deductive in multiple iterations, and our entrepreneurial job as traders is to use both. I may repeatedly observe a phenomenon and inductively figure out the pattern, but I then have to deductively reason that it will occur again and apply that to make money. I may deduce that a situation is potentially profitable, but I will have to use inductive (statistical) methods to fashion the strategy by which I capitalize.
January 24th, 2008 on 2:56 pm
Bill, I agree that strategy creation requires both inductive and deductive processes. And, when standing on the shoulders of giants, you are inherently relying upon whatever process they undertook.
I think that segregating the methods of reason is really a way to create a framework for how to approach a problem - in other words just how to start it. And it is an entirely personal choice based upon one’s preferences and past experiences. My favorite example of deductive reasoning is in over-corrections - I figure people’s characteristics don’t change, as you mention, so when bad news hits, people drop the security. So, I troll for headlines of bad events happening to decent or good companies. Funny enough, I almost pulled the trigger on CFTC calls right before the BAC announcement, but I avoided it out of my own fear!
However, one could easily approach it the same way… simply look for long-term patterns and stumble upon 30% drops that later return to 100% levels. Either starting point can work.
As for HF trading… I think in large part you are right… low return strategies with high leverage… practically providing liquidity to the big boys who really need to clear trades and take positions. People talk about a possible “correction” in the number of hedge funds - that’s where I speculate it would happen if it materializes.
Thanks as always for stopping by.
January 24th, 2008 on 3:55 pm
No problem. I like this place.
I can buy starting, or tending to start, from a basically inductive or deductive POV as valid, but the ultimate solution, even if started from scratch, is a blend.
I can also buy the starting point as being dictated by our biases and preconceptions (shaped by our experiences).
Market-makers have occasionally gotten messed over in fast moving markets. So have the HF quants.