I am new to the board and found it very interesting. I am also
working on a ATS and would like to exchanges ideas.

Using book data is intriguing and may have the price predicting
capability that is needed for a successful system. I have the
following
questions:

- Is there some simple experiment that I can run to show that book
data
can help in making better short term decisions

- What is so special about trading ES - does it have some statistical
characteristics that makes it suitable

- Most of the strategies here rely on optimization of parameters, but
how do you know it will work in the future

- What kind of strategies work best for ES - trend-following or anti-
trend

- How many trades are made each day


Following is a summary of my system (still under development):

- I am monitoring the performance of the system by running it on 45
stocks/etfs
- All are highly liquid and within $50-$100 price range (e.g. IBM,
AAPL etc)
- No parameters - same algorithm runs on all
- I have been running the system for the past 3 months over 60,000
trades
- About 5 have been constant winners week to week and real money
could
have been made
- For 20 some money could have been made - performance varied week to
week
- For 15 some money would have been lost - performance varied week to
week
- For the remaining 5 - performance was bad most of the weeks
- Some weeks, 80% of them made money others weeks about 40%

Not sure how to interpret these results, any insight will be
appreciated

Following are the basic ideas of the system:
- Anti-trend strategy - shorts when a new upward trend is detected
and goes long when a new downward trend is detected
- It produces about 60% winning trades
- Average loss per trade is higher than the gain
- It makes on the average 20 trades
- Works on 1 min samples of bid/ask/last
- I am not using JBookTrader - I use my own interface to IB

The question to the experts on this board is whether book data
will help in confirming the strength of a trend at that moment -
suppose at
t(n) the algorithm decides to buy/sell will book data be able to say
that at t(n) the trend appears strong, so don't go against it for now
wait for t(n+1)?

Is book data helpful in predicting very short term trends (couple of
minutes) ? Or there are other ways to do it?
Book data is not readily available - any expert insight will be
greatly appreciated - suggestions about how to proceed will be even
better.



--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to