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

Re: Identifying Pivots (wad Dow Theory Applied)



PureBytes Links

Trading Reference Links

Some quick thoughts on pivot definition:

1) Pivots should be identified according to strength which is the number of bars
on either side of the pivot point which are less extreme than the pivot  e.g.
strength of 2 means that two bars on either side of the pivot are less extreme
than the pivot (exception for flats - see below). The greater the strength of a
pivot, the more important it is.

2) A higher compression e.g. weekly vs daily can be used in lieu of greater
strength to identify more important pivots.

3) Use of averages and other filters will inevitably lead to misidentification
of some pivots.

4) Virtually all pivot identification routines appear to be designed by
programmers rather than traders and therefor fail to properly identify
conditions which are critically important to proper pivot identification:

a) The bar to the immediate left of the leftmost bar checked for strength must
be checked for True High or True Low i.e. most pivot routines do not check for
gap conditions at the far left which would nullify a pivot.

b) Bars to the right of a pivot may be flat (equal to the pivot extreme) except
that the rightmost bar in a pivot must be less extreme than the pivot. Most
pivot routines fail to identify flats. Further, when a flat occurs in dynamic
updating with the rightmost bar equal to the preceding bars, the routine needs
to remember this fact and, when the next bar arrives, check to see if the new
bar is less extreme thereby validating the pivot some bars further back than the
"strength".

Earl

-----Original Message-----
From: Tom Nagle <tnagle@xxxxxxxxxxx>
To: RealTraders Discussion Group <realtraders@xxxxxxxxxxxxxx>
Date: Tuesday, July 21, 1998 2:42 AM
Subject: Gen: Dow Theory Applied


>This brings me to my first problem - what is a high (low)?
>I've asked many people but not received a satisfactory answer.  It seems
>everyone knows what they consider a high or a low when they see it...but
>they can't give a definition, and they look at me as if I'm nuts when I
>ask the question.
>For instance, I've seen it suggested that you can define highs & lows by
>using a moving average and the highest price achieved whilst above the
>MA is a local high and the lowest low reached whilst below the MA is a
>local low. Comparisons of highs & lows can be made thereafter.