Automated trading

[Trading1903]

There are so many indicators out there.

But has anyone try to automate any of it ?

[yamantaka]

Most of these indicators have been automated in one form or another but abandoned because there was no edge in the indicators let alone automating them.

[CashManic]

Trading1903. Automated trading is THE ultimate goal. Then I can go surfing at my fav Bay area beaches and make moolah at the same time. Yup hands free trading is the ultimate dream. Automated strategies abound, but truth be told, they are rather mediocre. Eventually the answer is to invest a bit of coin in discussion and negotiation with a highly experienced NT coder. No doubt it will be a long drawn process but worth every cent and minute should the end product manifest to YOUR expectation and standards.

On the other hand,like you said there are a few quality indicators that really ought to be used in an autotrader strategy. Couple of problems though. Some vendors have restricted their indicators strictly to charts only and do not allow their scripts to be used in a strategy so they will not show up in the Condition Builder.It is becoming increasingly apparent nowadays that this is happening more often. You wanna use their indis in a custom strat? You have to negotiate and pay them extra leasing fees which definitely effs up your bottom line.

Do this. If you have some fav indicators you are thinking of using in an automated strat, check to see if they are ALSO in the indicator list of the Condition Builder of the Strategy Wizard first. If they are you're in luck and they can be used in strategies, alerts ,etc. Let us know if you come up wih some gems \\m/

[admis]

...there are a few quality indicators that really ought to be used in an autotrader strategy.

...Some vendors have restricted their indicators strictly to charts only and do not allow their scripts to be used in a strategy...

 

Hi CashManic, which indicators you mean?

[CashManic]

Hi admis. Here's a couple of them; Blue Wave ,Kwik Pop. Hawkeye indicators, Indicator Warehouse Diversified Trading Systems DTS, Triple Threat university t2u, Trend Following Trades tft, School of Trade JJ, Strategic Daytrading Jam/SDT, etc.

They DO NOT allow their scripts/indicators to be used in a customised autotrader .Now Kwik-Pop has their own quirky autotrader BBB but christ if they expect us to put it on a real money account.[-O<

Edited November 7, 2013 by CashManic

[mcl818]

Hi admis. Here's a couple of them; Blue Wave ,Kwik Pop. Hawkeye indicators, Indicator Warehouse Diversified Trading Systems DTS, Triple Threat university t2u, Trend Following Trades tft, School of Trade JJ, Strategic Daytrading Jam/SDT, etc.

They DO NOT allow their scripts/indicators to be used in a customised autotrader .Now Kwik-Pop has their own quirky autotrader BBB but christ if they expect us to put it on a real money account.[-O<

 

thank cashmanic for your infor

[Trading1903]

Thank you guys for your input. I do believe that the ultimate ( if there is such a thing ) auto trader should have its own artificial intelligence. It should "learn" as it goes. Even if you find an indicator that has a 60% + event IMO you need an AI. How do you create something like that is beyond me, but boy wouldn't it be sweet. I do not think such a thing can be created by one person. Perhaps a team. Like you I am still at the indicator stage, even with 8 plus years of positive discretionary trading experience ....

If anybody is interested perhaps we can start a one, I can put my experience and knowledge and when it comes to it $$. Everybody needs to contribute something. My problem starts with a programmer. I have spend money hiring , but even then it has not worked out as well as I expected.

[iksak]

From my experience (but I hope that someone can disproves me)there isn't a winning trading system working alone. People that works with trading systems has:

1- big capital to differentiate investiment;

2- at leas 10 different trading systems (but usually more) working on different instruments/time frames.

3- A monitoring system (VaR based or Standard Deviation based) indicating that a market behavior is changing, so you can quickly swich off a system (don't exist a system working forever, when the market changes the systems crash).

When you have different trading systems on different time frames on different instruments you can calculate your portfolio results (profit factor,drowdown, net profit,.....)and you must evaluate a de-correlation factor between profits and loses of your systems (when a system goes into drawdown the others usually must gain).

I also like the idea of going to the beach while my system is running but unfortunately it is not so. I want to say that IT IS VERY VERY DIFFICULT WORKS WITH FEW SYSTEMS AND FEW INSTRUMENTS and I think it is impossible working with only one!

But if someone wants to try some good idea, I will help you as I can.

Sorry for my english

[Trading1903]

Great points iksak. I do totally agree. Portfolio approach is the way to go. But where and how do we start?

[iksak]

IMHO the best way for testing portfolio systems (such as a SP500 securities rotation) is Amibroker. His programming language (AFL) isn't so easy but there is a great help guide. You can take a basket of securities and start to play with code (for instance buy the 20 securities with the best monthly or weekly performances and so on)