Good 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
LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Good system...

Post by LeviF » Wed Feb 06, 2008 9:10 pm

I was wondering if someone could post their test results of a "good" system that they created. I would simply like to compare the results to some tests that I have run. I'm not interested in copying the parameters or the system, I just want to look at the results to see if i'm on the right track.

Levi

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Thu Feb 07, 2008 5:56 am

When you searched, did you find these threads?

viewtopic.php?p=22877&highlight=intermediate#22877

viewtopic.php?p=22996&highlight=%2Acube%2A#22996

What's your opinion, do you consider either of these to be "good" ?

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Thu Feb 07, 2008 6:25 am

Thanks, i'll check these out. I had seen Ted's before.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

good? I think so

Post by painless » Sat Feb 09, 2008 9:00 am

My recent test on USSDCHF over 20 years
is now removed pending error adjustment.
Oh well thats ok. I'm getting used to egg on my face.

DeanoT
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Sat Apr 15, 2006 7:37 am
Location: Melbourne, Australia

Post by DeanoT » Fri Feb 22, 2008 3:57 am

As others have already stated, a 'good' system is likely to mean different things to different people.

Some may find System One below to be an example of a 'good' system, while most would find it untradeable.

Most would prefer System Two, however some would even consider a 35% drawdown to be too large to be a good system.

What is a good system to you, depends on you.
Attachments
System One.jpg
System One.jpg (154.42 KiB) Viewed 5237 times
System Two.jpg
System Two.jpg (150.41 KiB) Viewed 5237 times

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Fri Feb 22, 2008 7:34 am

One and Two sure do look like they share the same set of entry and exit signals, and only differ in the position sizing. Scan down the column of #trades in each of the years, One and Two are virtually identical. Compare the total #trades, total #winners, % of winning months, virtually identical. Leading me to believe One and Two get the same set of entry and exit signals, and just make different decisions about what size positions to put on.

If these simulations are trading futures or forex, the (Round Turns Per Year Per Million Dollars) are certainly large! Your broker must love these systems. It's a shame the bottom part of the Trading Performance table was omitted from the images. It would have been amusing to see the total dollars lost to commissions and to slippage, from such furiously high turnover. Your broker (who receives the commissions) and your market-maker (who receives the slippage) would probably make a very good living from these systems.

DeanoT
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Sat Apr 15, 2006 7:37 am
Location: Melbourne, Australia

Post by DeanoT » Fri Feb 22, 2008 5:15 pm

Sluggo

The two systems are indeed the same, with different position sizes as you correctly assessed.

Regarding the round turns per million, the figure is misleading, as the system is backtested over futures and stocks, which distorts the figures.

It is a very long term trend following system, evidenced by the average win percent to average loss percent ratio.

Regarding transaction costs, I have re-attached the tables highlighting slippage and transaction costs, along with the input assumptions.

As can be seen, transaction costs are, in fact, relatively small in proportion to total profits for both systems.
Attachments
Transaction Costs.jpg
Transaction Costs.jpg (19.34 KiB) Viewed 5133 times
System One.jpg
System One.jpg (169.67 KiB) Viewed 5135 times
System Two.jpg
System Two.jpg (159.89 KiB) Viewed 5135 times

DeanoT
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Sat Apr 15, 2006 7:37 am
Location: Melbourne, Australia

Post by DeanoT » Fri Feb 22, 2008 5:31 pm

Also attached below is a sensitivity analysis on the slippage factor. Even at 100% slippage, the MAR holds on quite well.
Attachments
Slippage One.jpg
Slippage One.jpg (21.41 KiB) Viewed 5125 times
Slippage Two.jpg
Slippage Two.jpg (20.29 KiB) Viewed 5125 times

Post Reply