Universal Forex Inefficiencies: Only a Long Term Trend Following Phenomena?

For most of my avid readers it shouldn’t be a mystery that for the past few months I have been very interested and focused in the development of systems which tackle very universal market inefficiencies. Following on the steps of the fathers of technical analysis, the idea here is to develop strategies which can give an overall profitable result when traded in most if not all available market instruments. The idea is that such a system is by nature extremely robust and only extremely drastic changes in the market would make it go bust since even drastic changes on single currency pairs or regions are bound to have hedging consequences which will lead to loss of profit on some instances and gains on others.

After doing a lot of research on this field, today I want to share with you some of my finding and the curious – and yet expected fact – that inefficiencies that are “universal” seem to be found only on longer term time frames when capturing longer term movements. I am going to go through some of the research I have done, exposing some of my conclusions as well as the reasons why I believe this seems to be the case. You will see why longer term systems provide much more universal systems while short term systems always fail to deliver across a wide array of instruments (with the exact same parameters).

-

-

Certainly it is possible to develop systems that trade a “whole market”, this has been the way in which futures traders have developed their portfolios for years since it became obvious to them that the only solution to robust trading was to obtain a system which they could trade on “everything”, eliminating the “selection bias” which plagues regular single currency/single system forex strategies. This is not to say that this type of strategy has no merit but merely that the robustness of universal systems is much larger and therefore these systems are globally regarded as the best possible solution when robustness plays the most important role.

If you do your research on systems developed for the Forex market you will find out that there are almost no reports of such universal systems evaluated on this particular market. The reason why this is the case seems to be puzzling but this is the main reason why I have decided to dedicate some Currency Trader Magazine articles to the exposure of such strategies. What you will find out however is a whole bunch of systems developed to trade the futures market which overall don’t work so well “as is” on Forex trading but which will work to some extent after you put some thought into the differences between both markets and how you can “correct” the system to trade on Forex instruments.

Developing universal daily trading systems hasn’t been such a terrible challenge since almost anything which makes a little bit of sense seems to generate profits on a universal portfolio, especially systems that attempt to capture long term trends such as the 390 period moving average cross strategy shown above. This strategy was evaluated over 16 different currency pairs using EOD data during a period of 11 years with the above results. Evaluating the strategy on pairs with data from 1990-2011 also shows that the past 20 years of trading have been equally profitable. This strategy is an example of a very robust strategy which achieves success on a “whole market” portfolio. The sacrifice made for such robustness is clearly profitability, with the above strategy having an AMR to Max DD ratio of 0.3.

I then decided to get into the task of finding a universal strategy trading shorter movements either based on some type of “ranging” algorithm or on some “switch” which allowed me to profit from trends or ranges according to some changes in market behavior. Although I did develop some very interesting indicators in the process – you will read about them within the next few months – I wasn’t able to find a single system which could beat the above mentioned long term MA cross a universal evaluation. The conclusion here seems to be that the chief universal inefficiency is following a very long term trend (which can last from months to years) and therefore anything which doesn’t take full advantage of this seems to be doomed to fail as a universal solution.

Nonetheless a few ideas have also started to come into my mind regarding this problem. In particular I believe that a sort of “global correlation index” might be usable in order to capture shorter movements or to gauge the possibility of success of certain overall movements. For example you could trade a portfolio in which you went “long EUR” by going long on all EUR pairs when certain correlation started to arise while you would go “short USD” when certain correlation rose on the USD. This would be in line with attempting to follow trends on the USD or EUR indexes, something which is much easier to do than say, following a trend on the EUR/AUD. You could also use this correlation to attempt to take profits from “disorganized trading” when no evident correlation could be seen. Obviously all of these ideas are out of the scope of the current MT4 implementation and therefore their testing will be left to our currently in development C++ or FreePascal testers.

From all the research I have carried out on this subject I would have to say that long term trend following seems to be the “only” universal inefficiency because it seems to be the only thing which all different pairs tend to have in common. They do have the tendency to stick to trends while all ranging behavior,  outside-of-trend developments and short trends seem to be governed by other more random market factors. Right now it seems to me that the best approach to the development of universal strategies is the improvement on the capturing of these movements as well as the quick elimination of trades that will not yield successful outcomes. It seems that the philosophy of cut your losses short and let your profits run is the most important thing when attempting to devise a universal strategy.

If you would like to learn more about my work in automated trading and how you too can learn to trade and design your own algorithmic trading strategies based on sound and robust trading tactics please consider joining Asirikuy.com, a website filled with educational videos, trading systems, development and a sound, honest and transparent approach towards automated trading in general . I hope you enjoyed this article ! :o)

Print Friendly
You can leave a response, or trackback from your own site.

2 Responses to “Universal Forex Inefficiencies: Only a Long Term Trend Following Phenomena?”

  1. McDuck says:

    your points are probably also valid for the stock market. I once read from an old trader somewhere that from his experience only two type of people really make money in the long run: excellent stock pickers (Greenblatt, Buffet, etc) and people who seat in a long trend (Rogers and others). Translated to the long term in the forex market, this comment would lead to same conclusions as yours.

    Best.

  2. David says:

    Yes, I agree. At least in Forex anyway, long-term trend following seems to be the only way to go. Without an edge you couldn’t make money, a market is either random and has no edge or it has a directional bias (a trend) and this has an edge. The longer the time frame the more significant the trend.

    The only other thing that signalled a strong long-term trend other than a break of a significant high or low was a large movement i.e. this week closes 1.5% higher than the high of last week etc..

    I’ve tested more systems than I can care to think about and the few that work all have these things in common: http://www.myforexdot.org.uk/best-forex-trading-systems.html

Leave a Reply

Subscribe to RSS Feed Follow me on Twitter!