Algorithmic trading - The Finance Forums: Financial planning, debt help, loans, mortgages, retirement and 401k, investment, and general finance discussion forums

brian · #1 (**permalink**) 10-17-2006, 09:36 PM

It's been the case for some time that people have been able to see patterns in the stock market - general long-term trends, but also apparently chaotic patterns.

I think it's fair to say that algo chasing on the stock markets is already a developing industry - which is why I found this interesting: http://www.techandfinance.com/index....al_developers/

Apparently it's a plugin for the Microsoft .NET platform and it sounds really interesting - especially this:

Quote:

“The problem with algorithmic trading is that it is more of a strategy. From the vendor you get a ready-made algorithm so most systems are developed in-house using .NET.”

But the curious question is - isn't there going to be some kind of Heisenberg principle in terms of algorthmic trading? Meaning, isn't predicting the pattern in the first place, then acting on those predictions, going to be completely flawed when there are a significant number of trading houses all using algorithmic trading to chase the same predictions?

Or is it going to be the case that the human influence will always prevent such systems working on a level playing field?

More to the point, is this gravitating to a safe investing strategy if applied to general long-term trends, and that there will always be room for the innovative investor?

bangfor · #2 (**permalink**) 11-23-2006, 01:54 AM

Anyway it’s looked at there are times money will be made because of some decision, when there are also times money could be lost with a similar decision, and if the positive frequency is more when you look at the pattern that becomes a winning formula till there is a higher negative frequency. What the algorithm can do is since it is driven by a computer it could multiply the number of these formulas to the point where you might not have to use the same formula again, which means it’s not going to be different than a chess game where the computer has millions of moves, and no matter how many people use it chances are they’ll always have their own unique formula, but when it comes to making money, there are external factors that make the price of the stocks go up and down.

4silverstrea · #3 (**permalink**) 01-07-2007, 07:25 PM

Regarding your last line, absolutely agree with you bangfor. So many unpredictables can change a company. I go on instinct more than anything. When I don’t, I lose money!

4silverstrea · #4 (**permalink**) 01-07-2007, 07:28 PM

But it’s not just about instinct. I assess certain factors: news reports, trends, the management team of a company and whether they are competent or not etc. Some people go by charts but a chart cannot predict the future, only the past performance. I apply the intangibles I have learned as a graphic designer to playing the stock market. What I mean by intangibles are these three things: anticipation; interpretation; and communication. We use these principles in advertising all the time.

4silverstrea · #5 (**permalink**) 01-07-2007, 07:29 PM

One more thing about charts. One of the guys on another forum board I post on said when in doubt with a chart, always look at the fundamentals.