Good article. In a small way this article decribes how historical data can be used to forecast the near future. This is THE concept that makes me gravitate to TA . The second concept that pulls me to TA is the idea that tommorrows price "MOST" of the time is near yesterday's or day befores price. Not all the time but most of the time. Maybe not exactly as yesterdays price but close enough to make money.roguetrader said:
However, my view is somewhat different than say a pure chartist. While I recognize that chart patterns do repeat and they do tend to resolve at least some of the time in certain ways I still don't use them much. Why? Because they are still subjective. If the basis of a forecast is subjective doesn't that tend to make the forecast subjective? However, I still strongly believe in using past data to forecast the near future. The difference, I think, is that you remove the subjectivity when your forecast is based on the behaviour of "actual" numbers and not on an interpretation of a pattern even though that pattern was created by actual numbers. I can visually see a pattern that you might not see. Or you might see one that I might not see. Or we could both see a different pattern. We have seen enough posts on this thread to validate what I just said. If one looks back over the posts it is easy to see that several of us looking at the same pattern saw things differently. Now, in such a case, our forecast will be based upon what we see and what we have been taught should be a probable outcome of such a pattern. Is that not correct? However, forecasting with pure price data using only the numbers removes the subjectivity. Am I correct in this? If not please correct me. So, my point is: if we take pure price data (i.e.numbers) and devise a system that does alot of number crunching then perhaps that system can "spit" out a probable (when I say probable I mean at least 70% and above) forecast of what prices "should" do tommorow based upon these numbers.
pttrader
Last edited:
.
