single moving average system

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
Nick
Full Member
Full Member
Posts: 10
Joined: Tue Jun 13, 2006 12:50 pm

single moving average system

Post by Nick » Sat Jan 20, 2007 5:32 pm

Hi all, this is my first post to the forum. I read all the very constructive posts regularly and intend to change status to active participant from lurker.

I am new to systems trading and have been looking at a single ema system for some time and wonder if anyone has tested such system where you would go long/short if the closing price is above/below the ema with a 3xATR stop and risk 2-3% of account equity. Going through this in metastock, it seems that if the system uses a reasonably long ema > 200 days, this would be a profitable long term trend following system. Any thoughts on this will or matters/risks to be considered in such system will be greatly appreciated.

Thanks,
Nick

Forum Mgmnt
Roundtable Knight
Roundtable Knight
Posts: 1842
Joined: Tue Apr 15, 2003 11:02 am
Contact:

Post by Forum Mgmnt » Sat Jan 20, 2007 6:54 pm

There are an unlimited number of different ways to determine if the market may have started a trend.

A single moving average works, simply checking for the current price against the price some number of days ago works, two moving averages work, three moving averages work, four moving averages work, moving averages plus a volatility channel work, and I can probably come up with ten more ways while I am typing.

The basic concept of trend following is very simple so it is easy to find ways to enter a trade as a trend follower.

I believe there is far less difference between any of the above approaches than most people would have you believe. Especially anyone who sells systems.

- Forum Mgmnt

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Sun Jan 21, 2007 1:31 am

I found from testing that shorter single moving average system can do a nice job, espacially on Soybeans market

Nick
Full Member
Full Member
Posts: 10
Joined: Tue Jun 13, 2006 12:50 pm

Post by Nick » Sun Jan 21, 2007 2:32 am

Thanks for your reply c.f./Barli.

I find the more I test/research that simple systems can be profitable and perhaps easier to follow.

Can this be tested by modifying the DMA system code or would a new blox need to be written to test such system in TB?

cramnhoj
Contributing Member
Contributing Member
Posts: 8
Joined: Thu Jul 06, 2006 5:27 am

Post by cramnhoj » Sun Jan 21, 2007 7:18 am

Welcome to the forum.

As c.f. has already said, there isn't that much difference in entry whichever of the different moving average methods you use. You can easily see this by plotting the different MA indicators on a price chart. If you do this on historical data, at the start of a long trend you'll see that they simply generate their entry signals on different dates but generally enter price continuously trends.

One thing that does make a difference depends on how many days back you use to calculate the MAs. The lower the number, the earlier the entry in a trend, but the tradeoff is more whipsaws. If you use a bigger value you'll see less whipsaws, but you end up entering later in a trend and that you'll only profit on the big trends. Having more trades and whipsaws while using a lower value for the MAs also means there's more chances that you'll get slippage. On the other hand some people just find long term trend following systems to be too boring for them. In the end, it's a matter of finding your own sweetspot.

With regards to position size, just make sure you test to see what max drawdowns you can end up having with whatever percent you decide to risk per trade. You then decide whether or not you're willing to undergo drawdows of that magnitude, if so then continue on with that, if not then lower your risk per trade and test again to see if you're ok with that drawdown. Because in a trend following system it's practically inevitable that a drawdown happens. For people starting out, they just have to realize that it's a normal cycle in system trading, it does not mean that they're not good at it or that their system isn't working. Ed Seykota likens trading to breathing, you have to breathe out before you can breathe in again. In the same way you just have to experience a drawdown before you start profiting again. Since one cannot know whether or not a trade ends up profitable at the time an entry signal comes out, one has to open a position to find out. But as long as you take signals that has a positive expectancy there's a bigger chance that you'll have a profit instead of a loss in the long run. It then becomes simply a matter of position sizing to make sure you don't end up getting a drawdown that you can't crawl out of.

Hope this helps.

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Sun Jan 21, 2007 8:23 am

Hmm, so very well put c.f..

It may take a few years of trading, reading and extensive backtesting to come to understand this.

Many years of striving to acquire worldly knowledge can lead to the realisation that at heart life is very simple.

Post Reply