|
PureBytes Links
Trading Reference Links
|
>Do you mean interesting in the profit sense or academic?
Academic. I don't know anyone who successfully trades so-called "high
probability patterns."
>And what does one use for a measure of confidence? Would a test
>against randomness suffice?
Maybe. The point I tried to make was, there's more to likelihood of
a desired outcome than mere probability that some result is lower
or higher than some other result -- which is typically where people
stop when analyzing patterns. Does the distribution of results have
a high standard deviation? Is it bimodal? Is the sample size of
qualifying patterns large enough to be statistically meaningful?
(The answer to that last question, in my experience, is usually
"no").
I'm leery of probabilities that turn out to be equal to the
fractions like 3/4, 4/5, etc. That may mean the sample size was 4
or 5... which isn't meaningful for statistics.
-Alex
|