Trading Discussion on Advanced Topics

Hey,


This is for those of you who have some ideas that are just out there and need to iron them out a little.

let me give some background on myself

I am a computational mathematics student at Arizona State University I have been demo trading and live trading for a combined 4 years. I have settled on a system and money management style I really like, I am getting ready to fully automate after another 3 months of seeing how it performs.

I have a team of 2 other asu students, one of them is a forex trader and the other is a programmer. We are currently ironing out a few different indicators we programmed and built.


My current topic and research. CLASSIFYING MARKET CONDITIONS.

I began my search by saying to myself if I believed in one system trading all markets ever. I don't know if it exists and I'll put money on the fact that I wont come up with it. So from there I began putting down facts about markets that may change but stay rather consistent across many markets. So I came out with an easy breakout strategy that trades the London and now I am working more with volatility. Daily pip movement, and trying to plot the values out on a normal distribution vs time. in other words at the start of the new york session i know 75% of the time I am going to get a total movement from a to b of x pips and total range of this and bla bla bla. if you plot these values vs time I think you begin to see breakout's coming



I am sorry for the crude explanation, what I really also wanted to talk about was if anyone has heard of, programmed, or seen neural networks being used for pattern recognition and trading. I have seen several people make "market meters" from pattern recogition and also statistical analysis of the likelihood of increases and decreases in price as price continues.

p.s I didn't want to be very detailed if I wasn't going to get any responses or any audience, which I am sure you can understand. If we generate interests we can go over exact process, math, execution, trading, or just theory in general...

I'm trying a similar strategy, although i'm not a maths guru, but I like the fact if you can get the right probabilities and stats you can become profitable so will be interested in following your thread.
A little about my strategy, that I'm only paper trading.
1. First component is to get average daily range of two pairs EURUSD & GBPUSD. currently both approx. 150 pips.
2. I get the High & low from 2pm - 8pm (Aust EST) and enter in spreadsheet which will calculate estimated high and low for the pending trading day.
3. I put in the close of 8pm and from there my spreadsheet will calculate whether a long / short or both ways can be traded on breakout. If the close is less than 80 pips from estimated high or low, the trade in that direction will not be taken. I put in a 20 pip breakout from the closing price.
4. Profit target is 50 pips, stop is 25, 1:2 risk / reward.
5. I repeat for 9pm bar also.
For example, if my calcs from step 3 & 4 show the estimated daily high at 1.3850 and low at 1.3700, and close is 1.3800, then my spreadsheet will say SHORT only, with an entry price of 1.3780.
Hope this makes sense.

Please note: This is trading 1 & 2 hours after London open.

should be right up your alley...
http://intelligenttradingtech.blogspot.com/