Trend Following: Profitable Reality or an Illusion doomed to

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.
marriot
Roundtable Knight
Roundtable Knight
Posts: 347
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Thu Sep 26, 2013 5:28 pm

Hi Roger,
your simulation looks really good.
Would you like to say something about used portfolio ?

Hi Anthony,
keep allocation low, leverage low is boring : )
Last edited by marriot on Fri Sep 27, 2013 2:15 am, edited 1 time in total.

Roger Rines
Roundtable Knight
Roundtable Knight
Posts: 1946
Joined: Wed Oct 06, 2004 10:52 am
Location: San Jose, CA

Post by Roger Rines » Thu Sep 26, 2013 6:03 pm

marriot wrote:Hi Roger,
your simolation looks really good.
Would you like to say something about used portfolio ?
Portfolio is a mixture of a 101 World Futures comprised of these groups:

Code: Select all

1: Currency Instrument Count:17,
2: Index Instrument Count:24,
3: Energy Instrument Count:10,
4: Soft Instrument Count:8,
5: Financial Instrument Count:14,
6: Metals-Prec Instrument Count:6,
7: Forrest Instrument Count:2,
8: Grain Instrument Count:15,
9: Metals-Ind. Instrument Count:2,
10: Meat Instrument Count:3,
11: Blank Instrument Count:0,
GetFieldCount(sGroupList)       ,11,
Symbol Count                    ,101, 

ecritt
Roundtable Knight
Roundtable Knight
Posts: 134
Joined: Sat Aug 28, 2004 3:44 am
Location: Phoenix, AZ
Contact:

Post by ecritt » Thu Sep 26, 2013 6:56 pm

I find it helpful to ask a related question. Assuming a guaranteed winning system (long-term), how likely is it from random chance alone, that at some point during a 30 year experience, a drawdown will occur that is both double the magnitude and double the duration of what is typical?

Consider the following hypothetical monthly returns:

-4.00% -3.10% -2.20% -1.30% -0.40% 0.50% 1.40% 2.30% 3.20% 4.10% 5.00% 5.90%

The annualized return is 11.38%.
The annualized volatility is 10.76%.
The drawdown is...dependent upon the path traveled.

If you randomly sample from this series of monthly returns there is virtually no way you can lose money over time. Yet, if you create hundreds of hypothetical 30 year track records...you'll see that the majority of them have one randomly placed drawdown that is much deeper and/or longer than all the other drawdowns. In these hypothetical scenarios it can't be because of high frequency trading, or the Fed, or too many trend followers, or finacialization of commodities.

I'm not claiming the markets haven't changed. But, I am saying we should expect to experience something even worse than the current environment over time, due to random chance alone.

Realistic expectations require an understanding of, and respect for, randomness.

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

Post by sluggo » Fri Sep 27, 2013 8:27 am

I hope you remember that Trading Blox has a feature called Monte Carlo analysis which does exactly this.
Image
Image
Image
Image

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

Post by AFJ Garner » Sat Sep 28, 2013 4:58 am

I second Eric and Sluggo's post. I also highly recommend TB users take a look in the Blox Market Place and down load the excellent random portfolio plug in. I recently ran a couple of thousand random portfolio tests using a balanced 25 instrument portfolio with no overall or sector risk constraints. Very similar results to Monet Carlo analysis. Doesn't really matter what system you use (I'm more interested in trade length: TF systems are all so similar anyway). I happened to use the BBBO on 80/2/0 with no profit taking and a 200 day long term filter.

http://tradersplace.net/forum/thread/22 ... o-survive/

http://tradersplace.net/blogs/60/bollin ... following/

I do, very firmly believe markets have changed and that TF has become less easy. In my mind there is no doubt whatsoever. I take trend efficiency deterioration as very clear evidence of that:

http://tradersplace.net/forum/thread/44 ... 380358505/

However, having recovered from being Corzined and generally recovered my equilibrium from the horrible event, I am convinced that TF will recover and prosper. Lessons have been learnt. In my own case the major lesson is to avoid greed and to keep leverage down. Period. TF is still a very good and very basic way to trade. Look at it this way, if TF does not work, then nor do indices.

The S&P 500 and every other such index out there is a trend following system.

Soph
Roundtable Fellow
Roundtable Fellow
Posts: 64
Joined: Tue Dec 02, 2008 12:22 pm

Post by Soph » Wed Oct 02, 2013 9:33 am

This is a bit more of an old school approach. However when you lay out 30+ year charts and examine them certain patterns tend to appear in how markets have changed recently. when looking 20-30 years back we see trend changes that, on a % basis are actually larger than what we see today. These trend changes are less frequent and, usually, less dramatic in their onset or demise.

More recent trends exist and are very evident, and often appear quite strong in their overall structure. However we see more frequent, and more dramatic, divergences from the underlying trend before the trend reestablishes itself. The breadth of the move is usually quite a bit less on a % basis as was common decades ago, however the move into it and out of it is more violent, and frequent.

A few important caveats of this very rudimentary observation:

1) a few categories of futures do not show this pattern. Bonds, as an example, due to the debt super-cycle we're in, and
2) one needs to not focus on the '08 event when examining the structure of overall trends.

So I would certainly agree that how we trend is much different. The trends are there still, but they are structured differently. Less trendy is as fair a definition as any.

ptan54
Full Member
Full Member
Posts: 11
Joined: Sun Sep 01, 2013 8:50 am

Re: Trend Following: Profitable Reality or an Illusion doomed to

Post by ptan54 » Wed May 27, 2015 10:59 pm

Re-reading this thread as of May 2015; some of the statements above came out to be prescient - after a bad period from 2009-2013, TF came back strongly in 2014.

We will have bad, horrible trendless years - keep the risk per trade small by assuming your drawdown will be 1.5X or 2X the backtested result. Most people only have enough capital (or investor patience, before redemptions force you to close) to do this for 5 years. If you were unlucky to start trading in Jan 2009 and used 1990-2008 as your backtesting period, you would probably have had a terrible real time trading experience from Jan 2009 - June 2014.

HWG
Roundtable Fellow
Roundtable Fellow
Posts: 84
Joined: Fri Jun 04, 2010 10:21 am
Location: Omaha, NE
Contact:

Re: Trend Following: Profitable Reality or an Illusion doomed to

Post by HWG » Tue Jun 02, 2015 3:04 pm

Funny that.

I was just reading this post and checking the date of all the replies and wondered how everyone felt now.

I "discovered" trend following around 2009 and have been toying with it on a part-time basis ever since but struggled. I thought it was because of me but it's pretty clear it was just bad timing. Good news is I haven't lost any really big money chasing it. At least not yet. :?

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

Re: Trend Following: Profitable Reality or an Illusion doomed to

Post by Chris67 » Thu Jun 04, 2015 6:57 am

2011-2013 was extremely tough - dare i say the worst period in history fo rmany that saw many long established trend following funds fall by the wayside
2014 was a blow away year - we recorded about 45%
2015 is pretty flat to date
I think the only conclusion you can reach is nothing ever really changes but draw downs and draw down lengths are always longer in the future than they were in the past

Post Reply