[UrmaBlume]

Long term trading success is not about instinct, divine inspiration or spontaneous intellectual combustion. It is about intelligent data processing and sound method.

If you are adept at handling, exporting and normalizing data there are a wide range of tools that can provide precise answers to questions such as those posed here - these are the tools the big boys use.

The right tools for the right job can make a lot of difference. There are many tools that can intelligently process huge amounts of data to discover precise rules, correlations and feasibilities in minutes that would take a manual processor decades to uncover.

In this case the most appropriate tool might be a rules generator, a decision tree or a Bayesian network. By the same token, given data and sample space, a genetically optimized neural network might produce a more effective, more granular output.

The practical application of such tools is discussed here, here and here. Those posts provide the names and sources of specific tools that may be of use and a discussion of how they are usually applied.

In the markets as in many games

Information = Equity

and in the markets the guy with the best information is the guy who makes the most money and he is also the guy who knows how to best process raw data into useful information.

If you are not processing at this level then you are not really competing at any level. It's not so much a question of whether you are using these exact same tools it is a question of whether your processing includes this depth of design.

cheers

UrmaBlume

[Tradewinds]

Originally Posted by UrmaBlume »

Long term trading success is not about instinct, divine inspiration or spontaneous intellectual combustion. It is about intelligent data processing and sound method.
UrmaBlume

I agree. I know nothing about MARS® (Multivariate Adaptive Regression Splines) or Neural Networks. It's interesting, and I have no doubt that it is very valuable. I have programed predictive curves using the Least Squares Method based on price. I haven't tried doing that with a combination of other inputs. But now you have got my brain cells vibrating a little bit here.

I doubt that I will ever use MARS® (Multivariate Adaptive Regression Splines) or Neural Networks. I guess I should never say never, but I'd need to look at the cost/benefit of getting that deep into the analytics. I'd be very interested to see performance comparisons of some big system compared to a similar methods and calculations in a good retail platform.

The data processing has to be based on something. A theory, conditions, a strategy, or observations. I'm not necessarily looking for a backtested condition that's been proven to a certain standard deviation. I'm all for that, and I hope to prove my system by that means eventually. But I'm just looking for the ideas that work in trading.

[Tradewinds]

Originally Posted by UrmaBlume »

If you are not processing at this level then you are not really competing at any level.

I now know that you are deep into cutting edge scientific methods of data analysis, so I'm assuming that you have references and data sources to back that statement up. Sorry, man. I just couldn't resist the temptation. I hope you take it in the light hearted way that I mean it.

[peterjerome]

a drawdown is a drawdown...it's either open(stay in trade) or closed (exit trade)
peter.