[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: pattern recognition


  • To: metastock@xxxxxxxxxxxxx
  • Subject: Re: pattern recognition
  • From: "rudolf stricker" <rst@xxxxxxxxxxx>
  • Date: Sat, 1 May 1999 05:43:20 -0400 (EDT)
  • In-reply-to: <3722DA58.2412DA67@xxxxxxxxxxxxx>

PureBytes Links

Trading Reference Links

On Sun, 25 Apr 1999 05:03:20 -0400, you wrote:

>has anyone been able to come up with a metastock formula for certain
>pattern recognition: head & shoulder, triangle ?

Are you looking for these "classical" patterns only, or would you
accept also more general "patterns"?

BTW1: Imo, all TA stuff is based on some kind of patterns (, but often
the patterns handled are not defined explicitly).

BTW2: Because TA stuff in most cases works only "on-dimensional", i.e.
it "looks" at is own history only, it can only be based on rather
simple patterns.

BTW3: There are many techniques (e.g. from statistics, fuzzy logic,
neural network technology, etc) that can handle multi-dimensional
patterns successfully. 

BTW4: The „Efficient Market Hypothesis“ is not generally valid.

mfg rudolf stricker
| Disclaimer: The views of this user are strictly his own.