Profit Target exits improve this system's performance

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

Post by danZman » Wed Jan 05, 2011 9:30 pm

LeviF wrote:For me, the conflict is that getting out of 99% of the position at 1R (leaving 1% in so the system doesnt jump back in the trade) is better than letting it ride. Again, this goes against all the theory of why LTTF works: returns arent normally distributed - you need to capture the 20+R fat tails, most people arent disciplined enough to cut losers and run winners or deal with lots of losing trades, yada yada. So then what is my edge?...
I feel like I just found out the earth is not flat and the moon is not made of cheese. I'm having trouble wrapping my head around this idea. Maybe that's why it works.

I will note that one of my systems (an intraday volatility BO method on ES), price targets do not work in any capacity. Time based targets DO work though.

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

Post by danZman » Wed Jan 05, 2011 9:51 pm

Just a quick thought...

Perhaps the fat tail gains are followed by fat tail pullbacks. Meaning
that many trends end in dramatic fashion (the pullbacks from the
peak are large). Sometimes those pullbacks come after a monster
move, sometimes at the start of a trend.

Sounds like some testing is in order.

rajivm
Roundtable Knight
Roundtable Knight
Posts: 129
Joined: Fri Jul 25, 2003 12:50 am
Location: New Delhi, India

What happens if your systems Add Units to successful Trade

Post by rajivm » Thu Jan 06, 2011 1:18 am

Hello Everyone,
A Good trendfollowing system will add units to successful trade so that you have a large position when you are right. If we do that in a System then maybe profit targets might not remain a good idea..Maybe

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Thu Jan 06, 2011 2:23 am

I think Sluggo and AF Garner are absolutely correct in their thoughts but just to add a small note of caution about standing on the shoulders of our predecessors - we still dont have anough data to say that markets have changed or that they will not revert back to the monster trends of the 80's - especially if the World is moving back into an inflatonary environment - it may well be that for the last 10 years taking profts at various points was very valuable - it may well be that for the next 10 years loading the boat and looking for 30 ATR moves is the best solution - we have no hindsight on this#
One thing I have notices in the last 6 months - with particular regards to Silver - is that its a monster move - and it has tried several times recently to collapse and take out all the long term system positions - but it isnt succeding - the trend is still in tact
I remeber in May 2006 SUgar and silver had made Huge moves up and they both reversed on a mega scale and took my long term positions out of the market in 2 days - moves I had been in for months were all given back in 24-48 hours - this time round I still find myself long sugar and silver as well as many other markets - despite some ugly pull backs something feels slightly different on this one - I cannot help personaly believing that as we enter 2011 the risks are that some of these moves keep on going - a lot of people are looking for the big pukes in commodities - what happens if by April we see silver at 4000, crude at 110 and sugar at 40 ?
Who knows but a combination of LTTF and what other members are suggesting in this post does seem to offer the best solution ??

all IMHO
Best
C

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

Post by danZman » Thu Jan 06, 2011 3:37 am

Chris,

Remindes me of Sluggo's post awhile back where he asks if the "Robusti family" of traders should look at all data, or a subset of data. link

Clearly (ok, not clearly, but the data doesn't lie) what they've shown is that if you use all the data, taking profits with a multi-instrument portfolio seems to be the way to go for the long run. For the systems they studied. Maybe in 100 years, trend following to the end will be proven. Maybe my grandkids can prove that one.

Still trying to wrap my head around this.

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

Re: What happens if your systems Add Units to successful Tra

Post by LeviF » Thu Jan 06, 2011 9:43 am

rajivm wrote:Hello Everyone,
A Good trendfollowing system will add units to successful trade so that you have a large position when you are right.
I dont think i've seen any systems (run over the past 20 years) that benefit from adding units.

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

Re: What happens if your systems Add Units to successful Tra

Post by sluggo » Thu Jan 06, 2011 10:29 am

I dont think i've seen any systems (run over the past 20 years) that benefit from adding units.
I have. And I'm trading them today with real money. (LINK)

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Thu Jan 06, 2011 11:03 am

Whether technique "X" creates better risk-adjusted return than a system without "X" depends on what other techniques are being used in the system.

For example, if you don't do anything with portfolio correlation then scaling-in can improve results in some circumstances if it happens to scale-in to trades that turn into the biggest winners on average. However, if you have portfolio correlation constraint rules in your system then scaling-in does not normally improve results since it's adding a 100% correlated unit to something that's already in the portfolio.

This "it depends" effect also applies to profit-taking rules. If your method relies on lots of relatively small winners to make money then a profit-taking exit at a relatively small R value will improve results; if it relies on a handful of large winners then cutting any profitable trades short will not improve results.

As an aside, profit-taking exits were the very last component of my trading system development and were only added relatively recently - I suggest traders think about the rate of profit of a trade as a potential way to exit winning trades early rather than an absolute R value or $ value etc.

Hope this helps

Paul

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

Post by LeviF » Thu Jan 06, 2011 11:53 am

Rate of profit is an interesting concept. How do you like to measure it?

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Thu Jan 06, 2011 12:02 pm

rate of movement=distance/time

rate of profit=?/?; you work it out ;-)

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

Post by LeviF » Thu Jan 06, 2011 12:08 pm

Lol. I know. These sort of concepts sound simple, but once you get programming them there are lots and lots of variables/parameters to consider that you never thought of initially.

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Thu Jan 06, 2011 12:13 pm

"Everything should be as simple as possible but not one bit simpler" as Einstein said.

Also, just telling you "this is exactly how I use rate of profit as an exit strategy" will not really help you determine how to effectively do it for your particular system(s) - hopefully it's just sparked a couple of ideas you can check out using TBB.

Paul

rajivm
Roundtable Knight
Roundtable Knight
Posts: 129
Joined: Fri Jul 25, 2003 12:50 am
Location: New Delhi, India

Re: What happens if your systems Add Units to successful Tra

Post by rajivm » Sat Jan 08, 2011 6:47 am

LeviF wrote:
rajivm wrote:Hello Everyone,
A Good trendfollowing system will add units to successful trade so that you have a large position when you are right.
I dont think i've seen any systems (run over the past 20 years) that benefit from adding units.
Hi Levi,
I use a system that adds upto 6 units in Long direction and no units in short. But I trade stocks only.

cryder
Roundtable Knight
Roundtable Knight
Posts: 101
Joined: Tue Jul 22, 2008 5:02 am
Location: Melbourne, Australia

Post by cryder » Sat Jan 08, 2011 4:39 pm

I also trade a (medium term) long only system on stocks that pyramids and lightens position size throughout the holding period. Holds winners for about 80 days and dumps losers in about 30 days. Win rate is (currently) in the 30 to 40% range.

Its a dual screen system that uses momentum type indicators to establish the longer term trend and the same momentum indicator with shorter time frame parameters to add on dips and lighten on peaks.

Has limits of a maximum of 3 pyramids and 3 lightens per position throughout the life of the trade. Been trading the system for 2 years to good effect.

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

Post by sluggo » Sat Feb 26, 2011 11:04 am

LeviF wrote:After more tinkering, i'm learning that taking off even more (ie - most) of my position at 1R is even better. It allows me to increase my bet size and get better risk adjusted returns. But I dont like it from a fundamental viewpoint. Everything I've been taught says I'm supposed to embrace and profit from the multi-year trends and not cut profits short... :cry:
While looking up a quote about retracements in Van Tharp's book, I stumbled across this passage. It made me smile:

"WHAT TO AVOID

"There is one kind of exit that is designed to get rid of losses, but it totally goes against the golden rule of trading, which, as we've said, is to cut your losses short and let your profits run. Instead, this exit produces large losses and small profits. It is one in which you enter the market with a large position size and then scale out with various exits. For example, you might start with 300 shares and sell 100 of them (when your trailing stop rises to your entry price)*. You might then sell another 100 shares at a $500 profit and keep the last 100 shares for a huge profit. Short-term traders use this type of strategy frequently. On a gut level, this sort of trading makes sense because you seem to be "insuring" your profits. But if you step back and really study it, you'll see how dangerous this type of trading is."

* I changed the wording within (parentheses) in a way that, I hope, clarifies Tharp's meaning
Last edited by sluggo on Sat Feb 26, 2011 11:09 am, edited 1 time in total.

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

Post by LeviF » Sat Feb 26, 2011 11:09 am

Yes. "Trust but verify" someone said. I'm becoming less and less of a trust-er the more I verify...

RedRock
Roundtable Knight
Roundtable Knight
Posts: 941
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Sat Feb 26, 2011 1:57 pm

Larry Williams "bailout" exit looked for a large move targeted exit on the day of entry or overnight but was willing to exit anywhere over break-even the next day while holding the original larger stop.

cryder
Roundtable Knight
Roundtable Knight
Posts: 101
Joined: Tue Jul 22, 2008 5:02 am
Location: Melbourne, Australia

Post by cryder » Sat Feb 26, 2011 10:29 pm

From Page 192 of Douglas, Trading in the Zone: " Taking Profits. Believe it or not, of all the skills one needs to learn to be a consistently successful trader, learning to take profits is probably the most difficult to master."

Douglas then goes on with an example similar to that quoted above, selling one third at breakeven etc; and the entire approach making the operator feel more comfortable.

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Post by stopsareforwimps » Sun Feb 27, 2011 9:00 pm

[sluggo] Students of Ralph Vince's book The Leverage Space Trading Model will recognize the above as nothing more than a straightforward application of his "Fundamental Equation of Trading" found on pages 50-54.
Jarrod Wilcox's book "Investing by the Numbers" has a slightly different formula which is fully derived in the book (p 96).

Expected (ln(1+r)) = ln (1+E) - 1/2V(r)/(1+E)^2 + 1/3S(r)V(r)^(3/2)/(1+e)^3 + other terms

Where

E = arithmetic average period (expected) return.

r = return. When we take expected ln (1+r) this gives us the log of the expected *compound* return.

V(r) = volatility of r

S(r) = skew of r[/list]

The second term shows why volatility is costly.

[Note I removed the material about Ralph Vince's formula - I tested it and it is roughly right though he provides no derivation I have been able to find.]

The third term is also interesting. It reflects the fact that positive skew is good, thus we let our winners run and cut our losses. Or at least we intend to do these things.

Academics reading this should know this is a Bermuda triangle and should Google scholar "NH Hakansson" before mentioning this in a paper.

http://scholar.google.com.au/scholar?hl ... =&as_vis=0

Hakansson trod on a few too many academic toes and made too many minor mistakes.

Tim Josling
Last edited by stopsareforwimps on Thu Mar 03, 2011 7:54 am, edited 1 time in total.

cliffg
Roundtable Fellow
Roundtable Fellow
Posts: 74
Joined: Wed Jan 11, 2006 5:18 pm
Location: Alberta, Canada

Post by cliffg » Sun Feb 27, 2011 10:25 pm

As an aside, profit-taking exits were the very last component of my trading system development and were only added relatively recently - I suggest traders think about the rate of profit of a trade as a potential way to exit winning trades early rather than an absolute R value or $ value etc. Paul King
Would a chandelier exit already encompass this idea? Seems like when the rate of change in p/l is rapid a chandelier reacts quicker. Is there such a thing as a dynamic chandelier?

I liked JD Canning's comments in the cotton discussion:
viewtopic.php?t=8009&highlight=chandelier
Rather than 'Sluggo's' remorse quandry, use a 'Canning'' non-emoting chandelier and focus on a method of reentry...

Post Reply