Systematic Trading research and development, with a flavour of Trend Following
Au.Tra.Sy blog – Automated trading System header image 2

Intricacies of Market and Trend Following Changes

March 10th, 2010 · 14 Comments · Strategies

nebulous intricacies

nebulous intricacies

In the last post we looked at the Turtle Trading system and saw that its performance went from outstanding for a long period of time to flat for 20 years. This opens a can of worms:

Does Trend Following work, is it dead, do markets change, does trend following rules need to adapt to these changes?

Let’s look at the different points of view.

The EMH Crowd

The EMH crowd does not believe in anything else than the Random Walk and by definition discards any profit-generating mechanical strategy.

As much as the Efficient Market Hypothesis (EMH) is a cornerstone of most modern financial theory, it has proven to be wrong partly because of some of its assumptions (not all actors in the market are rational, market prices are not fully random and normally distributed, etc.).

One great book that discredits the EMH approach and their descendant theories (CAPM, etc.) is by Mandelbrot: The (mis)behavior of the markets. It lays the arguments against the EMH in an approachable (ie not too much maths) way.

Trend Following is dead/does not work

Curtis Faith declared that “every few years trend following traders experience a period of losses and inevitably some expert will announce the end of trend following.”

Mike Covel also has a similar quote in his Trend Following book: “every 5 years some famous trader blows up and everyone declares trend following to be dead. Then 5 years later some famous trader blows up and everyone declares trend following to be dead, etc. ”

One of the main proponents of the argument against Trend Following is infamous trader Vic Niederhoffer (who “blew up” twice). He has been highly vocal about it, declaring Trend Following as one of the top Stock Market con.
Here is another link from his website to read more about it.

Arguments like this, despite the empirical evidence against it – in the form of Trend Following Wizards success, can be taken as a motivation for healthy skepticism and push you to strengthen your statistical research.

Are markets changing? (and must Trend Following change too?)

The self-professed “Trend Following poster boy” (a.k.a. Michael Covel) authoratively declares that this is a specious argument.

Occasionally, someone trying to promote something or start a debate will argue that trend following rules must always change due to changing market conditions. This is nonsense. It is a specious argument.

This is at best ambiguous. Covel likes to cite Bill Dunn who:

proffered that his basic system rules have not changed since 1974

Now, that is seducing: it seems to sell you the idea that you can develop a system, implement it and trade it for life. However this is not strictly true. As mentioned in this interview:

Dunn annually adjusts the parameters of trading signals and each markets weighting. In February – just as the grains were about to take off – he dumped the entire grain sector. But Dunn has no regrets.

Dunn is also known to have collaborated with Robert Pardo, a strong proponent of Walk-Forward testing (see below: a constant system adjustment).

To clarify: although Trend Following principles will never change, the rules/parameters of a Trend Following system might need to be adjusted to changing market conditions.

How can a Trend Following strategy adapt to the ever-changing markets?

In an Active Trader article, Anthony Garner attempts to discuss:

Do markets change? Is it necessary to undertake continued research and development and adapt a trend-following system to maintain its profitability over the years?

The article is only available for the magazine subscribers, but the result of the equity curve can be found on the Trading Blox forum. By “tuning up” the Turtle system, Garner manages to obtain interesting stats (MAR=2.26, CAGR=35.28%). The main change to the system is the use of a longer-term timeframe.

tunedturtle_139

Practically

I know I would not be happy trading the original Turtle System in the last 20 years and get a 0% return. If you started trading this system “back then”, when and how would you think it is time to switch to a revised system?

Let’s look at options:

1. Walk-Forward

Walk-Forward testing‘s principle is to keep running (in simulation) a “pool” of systems using different rules/parameters. At regular interval, you evaluate what systems are best (performance, robustness, etc.) and trade those until the next re-evaluation. The potential risk with this approach is that you might end up like a dog chasing your tail.

However, this approach would have you switched from the original Turtle system to the new one a while ago.

2. Alternative Walk-Forward

Markets exhibit some degree of inefficiency – and Trend Following is a strategy designed to profit from these inefficiencies. I am still refining my theoritical understanding and explanation of it, but I believe Trend Following’s performance is mostly the result of fat-tailed distributions (distribution kurtosis) and possibly autocorrelation.

If one can associate the evolution of these characteristics to the performance of Trend Following systems it might be possible to adapt the system rules/parameters to the values and evolution of the price distribution characteristics. This is a topic I’d like to investigate further.

3. Mixing Systems

Finally, and this seems to be a strategy adopted by many professionals: mix different systems and different timeframes. Here the rationale is that we cannot predict what systems are going to under/over perform and mixing several ones together will smooth out the equity curve.

Related Posts with Thumbnails

Tags: ······

14 Comments so far ↓

  • Joshua Ulrich

    Even after he accounts for market impact, Anthony Garner’s results are impressive (MAR=1.46, CAGR=21.82%).

    I don’t have the results in front of me, but I would bet that the original Turtle system beat Garner’s walk-forward system in the 1970′s/80′s, but that Garner’s system is much more consistent over the entire sample. As you said, you must carefully choose your walk-forward criteria to keep you from chasing the hot set of parameters.

  • Jez

    Hi Joshua,
    Yes – I think I’d settle for these results quite happily.
    I have not been able to read the original article yet (magazine is not sold in the UK and old articles are available online after a few months) – only the discussion in the Trading Blox forum. But I was under the impression that the system from Garner was static (ie not walk-forward)?
    Check the link (to Trading Blox forum) on my last post with original Turtle System (http://www.automated-trading-system.com/turtles-just-lucky/ ) the CAGR was 216% from 1970 to 1986!
    I think with a good Walk-Forward process, it “migh”t have allowed you to switch between different versions of the system as performance evolved…

  • Prashanth

    The underlying point is “Is building a Trend Following System Easy”. I believe its not. Most arguments against Trend Following assume simple systems and when they fail try to point out to the inadequacies of the system.

    Basically Trend Following is all about catching the trend as early as it starts, getting off as early as possible when it ends and having the system ignore as many as whipsaws as possible using filters. If you are able to achieve it, TF works for you, else it does not. Its as simple as that. Fact is that trends still continue to exist and the strategy will be to catch as much of the same as possible.

    Regarding Michael Covel, well the guy sells his products for a living. If he was really good at it, wouldn’t he be managing a fund by now. I am a regular visitor to his site since it provides me with info at a single place but his sales pitch every now and then does start to create irritation.

    The fact that a known Trend Follower David Harding employs over 100 Ph.d’s in Mathematics is proof enough that building systems using any strategy is not child’s play anymore. Its serious business for serious money.

    I am not sure you have given the correct meaning of Walk Forward testing. Basically Walk Forward testing deals with running a strategy that you have devised using one set of data on another set of data which has not been used for the optimization of the system that was build.

    For example, if you were to optimize a system based on say data between 1990 – 1995, you could run the same system using data from 1996 – 2000 to see how it could have performed in data not used for the test. By doing this, you can see whether your system is curver fitted (which will be reflected in excellent quality results in In Sample testing and terrible performance in Out of Sample testing.

  • Jez

    Prashant,
    Agree with you, building a Trend Following system is not as easy as it can sound (in some popular books) but I think you missed a very important point: Money and Risk management. I believe that it can have more impact to the overall performance than the system signals.

    With regards to Winton, I am not sure he just runs a 100% Trend Following fund. He definitely wants to build a firm to compete with Renaissance (with other strategies than TF) and that might be a reason for so many PhDs (as well as the possible marketing weight of it).

    And for walk-forward testing, please read the link to my older post explaining how it works (http://www.automated-trading-system.com/walk-forward-testing/ ). There is a cool animation illustrating it. What you are describing is simple out-of-sample testing whereas Walk-Forward is an adaptive process which re-optimises the system on a continuous basis.

  • Sam

    Guys it’s important to remember that even a system built on in sample data and successfully tested on out of sample data has a certain probability of falling apart at any given point in the future. This is due to the well known fact that a historical pattern has no obligation to remain intact for any given period of time.

    It is wise to always be skeptical of any kind of back tested results regardless of whether or not it was run on out of sample data or walked forward. Just because it has performed on these particular kinds of tests does not mean it is not curve fit.

    Things exist in the market until they don’t.

  • Troy S.

    A very interesting a relevant post (not that your other posts aren’t interesting and relevant too!).

    I especially like your alternative walk-forward idea. Determining a relationship b/w returns and fat-tailed distributions/autocorrelation, and its predictive value, would be another way of building a more adaptive system (in addition to your walk-forward method in #1, and the one you mention in the post above).

    Garner’s gorgeous equity curve almost made my brain explode. However… I may be dense and completely missing something, but it seems to me that Garner just tweaked the Turtle system with the benefit of hindsight, and not as a result of walk-forward testing?

  • Jez

    Great comments Troy and Sam, thanks!
    Completely agree on the skepticism in general and the possibility of the system breaking down after successful out-of-sample testing. If you think about it this can also be curve-fit: if you test/optimise 100 systems you will probably find one which performance will be very good (by effect of random luck). This is “1st degree curve-fit” if you like… If you now take 10,000 systems, you will probably find 100 very good ones in your optimisation. If you now test these 100 systems on out-of-sample data, you will probably find 1 very good one (random luck again). This is second degree curve-fit. The system is likely to break down in the future because randomness and survivorship bias only made it rise to the top… The same concept applies to any backtesting process. The trick is to limit the odds of picking a “random winner” instead of a true one..

    That being said, an appealing aspect of walk-forward is that it is a sort of dynamic backtesting process that should self-adapt. Fwiw I have no strong opinion for/against walk-forward.

    It is also true that Garner probably had the benefit of hindsight in tuning his turtle system although I have not found published the method he used to tweak the system (pretty sure it was not walk-forward). It could well be backtesting 100s of params and picking the best ones (although I know that he has generally been leaning towards the long-end of the trading timeframe).

    What excites me is the challenge of formalising/automating the optimisation/verification process so that it can be objective and devoid of as much curve-fitting as possible. Some challenge!

  • david varadi

    nice work jez and a good article. i would comment that there doesn’t need to be dividing camps between walk-forward testing and the robust long-term careful testing camp. effectively as humans we are evaluating systems in a way that can be automated, thus ideally the compromise is to simply create a walk-forward test that mimics what we as system developers would be doing at a given point in time.

    best
    david

  • Michael Covel

    Quick clarification. The “every five years trend following is declared dead” line is from page 20 of my book “Trend Following”. The source is not Faith. In fact, Faith’s only money management firm since the Turtle days did go bust and ended in a CFTC permanent ban.

    Also, in terms of the “are markets changing debate?” … I hold my (2) books out as my view on the subject, not one blog post alone (even though I support the post).

    Keep up the good debate guys! :-)

  • Jez

    @Mike – Thanks for dropping by.
    Credit where Credit is due: I have updated the post to reflect this – Apologies, as I did get my quotes mixed up between yours and Faith’s (in his Way of the Turtle book) which are very similar.

    Regarding the debate, and I did enjoy reading your books, I still came away with the impression that your writing made Trend Following sound easier/simpler than it actually is when digging in the details – although I cannot comment on the material in your courses (which might include some adaptability rules/systems similar to what Dunn might be doing?)

  • Michael Covel

    Jez, I am not sure what worthwhile endeavor in life is easy! That said, I have thought about giving up my trend following preaching now that I have read “Trading from Your Gut!” That might make a good book review here?

  • Jez

    very true and I agree: nothing’s easy… I was just saying that I wished your Trend Following book covered more of the not-so-simple nitty-gritty details.

    Ahaha, re: Trading from Your Gut: not much chance of a review here as I have no interest in this book (except to see how he can justify this radical shift from being a system trader)

  • Michael Covel

    Jez, not to belabor this post direction, but there is no radical shift with this one player. His bio is an open book now and ‘system trader’ was never an accurate description. Discretion all over — even going back to Turtles days. That’s not even getting into a myriad of other inconsistencies and controversies that cast doubt.

  • Trend Trader Jim

    “Now, that is seducing: it seems to sell you the idea that you can develop a system, implement it and trade it for life. However this is not strictly true. As mentioned in this interview:

    Dunn annually adjusts the parameters of trading signals and each markets weighting. In February – just as the grains were about to take off – he dumped the entire grain sector. But Dunn has no regrets.”

    This is not an example of changing the system. This is an example of changing a sector of the portfolio that the system trades… there is a HUGE difference.

Leave a Comment