Percent Profit Factor

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
shepherp
Contributor
Contributor
Posts: 2
Joined: Mon Apr 06, 2015 8:28 am

Percent Profit Factor

Post by shepherp » Tue Apr 28, 2015 8:54 am

Hi. What are the inherent flaws in using Percent Profit Factor to filter markets selection when creating my portfolio? I understand using past data no guide to future profit contribution etc but given that's the premise for testing all parameters using optimization anyway, Percent Profit Factor seems a very good starting point for selecting which markets to include in a large diversified futures portfolio. Thanks.

Tim Arnold
Site Admin
Site Admin
Posts: 8500
Joined: Tue Apr 06, 2004 1:41 pm
Location: Boston, MA
Contact:

Re: Percent Profit Factor

Post by Tim Arnold » Tue Apr 28, 2015 9:20 am

I understand using past data no guide to future profit contribution
Exactly. Test a basket of 50 markets from 1990-2000 and pick the top 10 based on some criteria like Percent Profit Factor. Then test these 10 markets from 2001-2010. Compare this to having traded all 50 from 2001-2010 or a random selection from 2001-2010.

sinkpretend
Full Member
Full Member
Posts: 21
Joined: Fri May 08, 2015 8:40 am

Re: Percent Profit Factor

Post by sinkpretend » Wed May 27, 2015 5:25 am

Its always considered as one of the best practices to analyze data before decision as Arnold indicated the range and steps. Here is a link which will help you to analyze data and understand more.

shepherp
Contributor
Contributor
Posts: 2
Joined: Mon Apr 06, 2015 8:28 am

Re: Percent Profit Factor

Post by shepherp » Wed May 27, 2015 5:36 am

Tim, Sinkpretend, thank you v much for the input; its helpful to feel more on the right path.

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

Re: Percent Profit Factor

Post by AFJ Garner » Wed May 27, 2015 4:36 pm

For what it is worth my belief is that cherry picking a portfolio is one of the best areas of self deception in the whole business. Random portfolio testing is one of the best ways to test a strategy. Easier with mutual funds or stocks, somewhat more restricting with futures where the available range is much smaller.

"Optimisation" seems to me to be a very dangerous concept but of course some weeding out has to be done in some areas such as the parameters of a system. No easy or lasting answer.

Post Reply