Hi. I’d like to ask for some advice: I want to use a month or two of fluctuations in a USD currency index to determine what numbers to use as triggers to automate trading. If, for instance, over the course of a month an increase of X in the index was followed by an increase of Y before it was followed by a decrease of Z 80% of the time, then I will automate the platform to buy when X happens and sell when Y or Z happens. Will this work? How many percent should I expect to make in a month this way?
When you say that I “will probably find the percentages are such that there’s no real worthwhile edge to be had,” what sort of percentage do you figure such a strategy would yield? My goal is not to achieve extremely high percentages, but to have trading profitably automated so that my attention is not generally required. So if it makes even a few percent per month, then it would be helpful.
What I really want to ascertain is whether fluctuations are consistent enough to allow such a strategy to work on any normal trading day.
What I mean is that you are likely to find that the distribution of outcomes following the Y move are such that you’ve got very close to even odds of either a continuation or a reverse and that the expected return of those trades is very near zero before taking in to account the spread and any other transaction costs.
If you mean by doing something like having a stop at X and a take profit at 2X then I’d say no because the probabilities of the two moves will offset. By that I mean the probability of getting stopped out would be greater than that of reaching your target to the point of canceling the larger gains relative to the losses out.
If there’s a basically balanced distribution of returns either side of zero (as you will most likely find), then it won’t matter what size your stops or losses are. Again, though, run some numbers so you can see what I’m talking about. Maybe along the way you find something interesting.