Trends Can Not Be Predicted
No trend can be predicted. Even though prediction is an impossible goal, many chase it every waking minute. Christopher Crude, from ©Alternative Investment Management Association, offered some gentle sarcasm and quick wit to the majority forever fixated on forecasting trends.
The thing about trends is that they are much easier to spot in retrospect and tantamount to impossible to recognize as they are beginning.
...as many investors and a number of members of the press know to their chagrin, a technically driven investment approach inevitably leaves the manager without a quotable view. His (or her) belief/expectation/notion that the euro, for example, will rise or fall against the Swiss franc may or may not be interesting cocktail party conversation but has no impact upon the way the computer will react.
I would prefer to finish with a certain currency forecast, based upon my own fundamental reading of the market and one which underpins my personal investment philosophy. This is that the dollar will inevitably trade at 0.5 euros and at 2.0 euros. The only problem is I can't tell you when this will happen or which event will be first. On that basis alone, it seems best to stay with our systematic approach.
Mark Rzepczynski of John W. Henry and Co. had this to say also:
We're a trend follower, okay, we use just price information and volatility in order to make decisions okay. The reason why we do that is because we don't think that we can predict the future okay. We don't think we can predict the future because we can't go back to our original decision tree example, we can't think of all the probabilities and all the possible outcomes. And because of that, what we have to do is take a different approach. Also because we're trading the markets and we're active in them everyday in our programs, we have to come up with a method in which we can be able to make fast decisions...I can't be an expert in everyone of them [markets]. In fact I can't be an expert in any of them, so what I have to do is be able to be expert at being able to move faster when I see information that's important...So my way in which I can move faster is to just use the price information that's the aggregation of everyone's expectation. Our signal is the trend information which is surrounded often times by noise. And so what we try to do is extract the appropriate signals as quickly as possible so we can act fast to limit our risk and also create opportunities. Now we're frugal in the senses that we use sort of some very simple recognition heuristics and we say that we have recognition heuristics we're looking at the price information itself. Now what could be an example of this...We like to think of those as non-linear models. But it's no different than what some people describe as breakout systems.