Interesting Equity Curve

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:

Interesting Equity Curve

Post by LeviF » Wed Dec 10, 2008 10:21 pm

This is a short term forex system that I'm working on. This is the equity curve from July 1991 to Jan 2008. Max DD of 9.3%, MAR of 1.7. Monte Carlo 95% DD of 15%.
Attachments
curve1.JPG
curve1.JPG (54.21 KiB) Viewed 5180 times

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

Post by LeviF » Wed Dec 10, 2008 10:24 pm

This is the curve including this year. Max DD to 20%, MAR to 0.7
Attachments
curve2.JPG
curve2.JPG (51.04 KiB) Viewed 5178 times

Demon

Post by Demon » Thu Dec 11, 2008 6:36 am

Hi Levi, I'm guessing you optimised this on the July 99 to January 08 data...

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

Post by LeviF » Thu Dec 11, 2008 8:09 am

Thats what it looks like, but no.

danZman
Roundtable Knight
Roundtable Knight
Posts: 117
Joined: Thu Mar 10, 2005 10:07 am
Location: Newport Beach,CA

Post by danZman » Fri Dec 12, 2008 12:14 am

If my equity curve changed that dramatically, I would find something else. You could always play the equity curve game though. Factor in a MA of the actually equity curve and stop trading it...and then restart when it comes back from reform school.

D

Asamat
Roundtable Knight
Roundtable Knight
Posts: 175
Joined: Fri Jun 03, 2005 7:50 am
Location: Walldorf, Germany

Post by Asamat » Sat Dec 13, 2008 11:52 am

Hi levijean,

this is a frightning demonstration (if indeed not curve fitted). Thanks for showing us. My 2 cent:

a) Actually you are still within the parameters of your backtest. You wrote: "Monte Carlo 95% DD of 15%", so the current 20% don't seem incompatible with that. Of course look and feel of the underwater curve tell a different story ...

b) Systems with an edge always exploit some market inefficiency. (Thereby they reduce that inefficiency and make the market as a whole more efficient.) It seems the inefficiency your system was designed to capture has disappeared, at least temporary ... I've seen that happen to other systems, but not like that, not like a switch.

I for one would be interested in how that picture looks in some months.

Regards,
Asamat

Asamat
Roundtable Knight
Roundtable Knight
Posts: 175
Joined: Fri Jun 03, 2005 7:50 am
Location: Walldorf, Germany

Post by Asamat » Tue Aug 25, 2009 12:08 pm

Hi levijean,

any update on this? Did you follow how the past eight months would have been?

Regards,
Asamat

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

Post by LeviF » Tue Aug 25, 2009 12:30 pm

The original curve was based on bad data and therefore flawed from the get-go.

twells5150
Full Member
Full Member
Posts: 15
Joined: Mon Jan 08, 2007 4:56 pm

Can you elaborate?

Post by twells5150 » Wed Sep 09, 2009 12:29 am

Can you elaborate on the "bad data".

Was it just for 2008 or the entire timeframe.

What does the graph look like now with the good data?

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

Post by LeviF » Wed Sep 09, 2009 12:49 am

I dont care to get into it. There is nothing to be learned from this experiment.

Post Reply