Developing a Trading Model
The stock market can only be understood backwards, but must be practiced forwards! And you need lots of practice.
---------------------------------------------------------------------------------
------------------------------------------------
To do any job well, to perform any skill with competence, you need good ideas -- a good model -- to guide you. That holds for playing the markets as much as it does for any other activity. But it's seldom clear precisely what you should do.
Maybe Richard Ney has it right, that the specialists control the markets and that we should do what they do. But how do we learn what they're up to? They certainly don't send us tips. I personally haven't heard from any of them. Have you? So you still need a way to see what's happening? Correct structuring of the situation dynamics still seems necessary. Good ideas become essential.
Anyone can make predictions, of course, and many of us do just that. But that's not to say their predictions are accurate. For accuracy there has to be a real unfolding recognizable pattern, plus you have to be able to SEE the pattern. For example, you can predict what will happen to you if you jump out of an airplane. You can predict it because we have a principle of gravity that says you will fall to the ground. Similarly, you can predict what the market will do if you can see an unfolding pattern in prices. It's possible using Elliott Wave Theory, for instance.
Whether you go with Ney, or with Elliott, you still need to keep your wealth where it will do the most good, whether it's in questionable US dollars, Chinese coal stocks, Australian land, Euro bonds, high risk tech stocks, or anything else. You need a guide to make your decisions -- to trade what you currently hold -- like dollars -- for something else.
We need to look to ourselves and apply trading guidelines. A model structures the setting, thereby generating information, and should let you recognize price patterns. The model is hard to find, though, and I've personally been leaning more on traditional price patterns and relying more on Elliott to do the reading.
---------------------------------------------
Let's face it -- things don't ever, or at least seldom, add up nice and neat in a clear model. The real world is far too complex for that to occur. The best we can really hope for is a good working idea or set of ideas.
Since the specialists' order books aren't available to us, our only option is to study the time series of prices and project the flow of money around the market environment -- like the flow of a tennis ball around the court. In tennis you decide where to hit the ball. In the market you decide where to put your money. In either event you need principles and mechanisms of behavior to produce good results.
Tushar S Chande believes building a model is essential, whereas J M W Tadion thinks it is practically impossible, and Ney says it plays into the hands of the floor specialists!
-----------------------------------------------
Intermarket Analysis looks at price charts of the world's markets and their interactions. In this age of globalization, it's clear that the markets are interconnected. For this you might find it useful to check Barton's Asian Stock Charts.
According to Tadion:
[It] is reasonable to expect that there must certainly be present [in the market] a number of fundamental forces that are responsible for the market's activity, for, by definition, the fundamentals of every process must determine the outcome of that process. But, as far as the market is concerned, things are not as simple as that, for to unravel and identify the multitudes of "fundamentals" that could conceivably influence the market, that is, the market participants, is extremely difficult, and to quantify them and analyze their impact and interaction is much more difficult still.
To try to make the job easier, we might get a jump-start with existing trading formulas, like those of Murray A Ruggiero. Even so, the formulas are only preliminary descriptions.
------------------------------------------------
One possible way around the modeling difficulties is to generate a representative sample of the markets by selecting securities and reading their charts to recognize tops or bottoms in individual markets -- money moving in or out.
We don't have formulas linking chart attributes quantitatively. At best they'd be implicit -- i.e., hidden in the estimates derived from your reading of the chart data. Implicit or explicit, though, they still have to be viewed against the backdrop of the charts, which, taken together, should let you keep your market orientation.
Charting techniques using traditional theory and wave theory, together with Fibonacci numbers, and indicators could give us a leg up on investing and might be the way to go. So also might be work on neural networks. Nevertheless, all of this could still be just a vain gambit. You still have to "climb the wall of worry" and avoid "falling on the slippery slope of hope."
------------------------------------------