Thanks for the reply, Bill. Like you mentioned, what makes this so hard is that TA is very subjective--it's all in what the chart reader interprets from the various signals, price movement, volume, etc. And what I am really asking here is what popular indicators or systems would be good to use for such a project.
But you are wrong in that I am searching for the Holy Grail of indicators. Whether they turn out to be useful in predicting movement or not, the purpose would be to find this out, one way or the other.
I am initially thinking that I will test various indicators by themselves, over different periods (such as 1 day, 5 days, and until opposite signal is given). How does this sound? I would also take a look at what happens when different indicators are either in agreement or disagreement... Then perform statistic analysis on the results, and come up with some kind of regression equation.
Also, I am guessing that numerous studies have already been done, on such a subject with such potentially huge financial ramifications in the computer trading business. Do you know of any resource that would list these studies or their results?
Thanks again,
Brian |
