I'll second Shogun's recommendation. A good broker for automation is Interactive Brokers and in that case you'll probably want Dynaorder for executing limit and stop orders and monitoring positions and cash balances. There are a couple examples of fully automated EFS scripts on the Dynaorder examples webpage.
BakedWafer,
Thank you for your reply. That makes a lot of sense and solves the problem I have been facing about the indicators getting affected by a big overnight gap, especially oscillators. How can I prevent this from happening? How can I can base the signals for the system on the current session only? Do I set the indicator to restart the calculation on the open of the session? Any thoughts you can share on how I can accomplish this would be greatly appreciated. Or did I misunderstand your recommendation?
[QUOTE]Originally posted by chrisf95003
[B] How can I can base the signals for the system on the current session only? Do I set the indicator to restart the calculation on the open of the session?
Chris,
I think you don’t have to do anything at all.
If you calculate for instance a 14 periods RSI you need 14 periods of correct data. This means that you have to wait for 14 periods starting from the open of the session to have the correct data. Even if you don’t tell your system to start calculating from the open of the sessions you will have to wait 14 periods to start trading. At that time the data from last session is 15 periods ago, thus not interfering in your calculations anymore. You will get exactly the same indicator as if you started counting from the opening of this session.
You are asking how to address the problems of GAPS in price or screwed up indicator actions from the end of one day to the next??
The only way to resolve this is to allow for extra bars at the beginning of each day (say 20~40 if necessary) to allow for some "normalization" of the indicators in question. This will resolve your problems with the indicators at the start of the trading day..
Now, to address your trading system, you need a StartTime and EndTime coded into your trading system to know when to allow the entry triggers to "go off"...
so you might have a starttime of 930 and an endtime of 1500. Thus, you would need to check for this within your trading system and allow the entries to go off only between these times.
Originally posted by Doji3333 You are asking how to address the problems of GAPS in price or screwed up indicator actions from the end of one day to the next??
The only way to resolve this is to allow for extra bars at the beginning of each day (say 20~40 if necessary) to allow for some "normalization" of the indicators in question. This will resolve your problems with the indicators at the start of the trading day..
Now, to address your trading system, you need a StartTime and EndTime coded into your trading system to know when to allow the entry triggers to "go off"...
so you might have a starttime of 930 and an endtime of 1500. Thus, you would need to check for this within your trading system and allow the entries to go off only between these times.
This is a client's system that is just kicking some *** at the moment.
This is a 120 day test with a "Running Equity" feature that I developed to compound equity at is runs the test. So it is starting with 1 contract and ending with about 30.
Was this the day trading system? Is the system is able to pyramid from 2 lots to 30 in one session? Is that normal for an automated sytem to get that agressive?
Originally posted by chrisf95003 Was this the day trading system? Is the system is able to pyramid from 2 lots to 30 in one session? Is that normal for an automated sytem to get that agressive?
Thanks,
Chris
Chris,
i don't think this system goes from 2 to 30 contracts in one session.
I think the system starts small and takes bigger positions as the equity grows . So after a few trades the equity should allow the trader to take bigger positions because he reinvests his profit.
I also think it is not an intraday trading system because the margins are at the level of overnight positions.
If have systems that perform much better in backtesting, but that is irrelevant. Only a track record of effective trades can proof the quality of a system, that's my final conclusion after 15 years of trading.
Comment