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

back to the well again... parallel trend lines



PureBytes Links

Trading Reference Links

John,

let me restate what seems to be the core of the problem you want to solve.
If it isn't, please let me know.

On a semi-log chart (regular time scale, logarithmic price scale), let a
straight line be defined by two points P1(x1,y1) and P2(x2,y2). Given a
third point P3(x3,y3) not incident to the original straight line, we are
looking for a point P4 such that, on the semi-log chart, the straight line
P3->P4 will be parallel to the trend line P1->P2.

Solution: P4(x3+x2-x1, y3*y2/y1)

Again, there are infinitely many other possible points P4.

Hope this helps.

Michael Suesserott


-----Ursprüngliche Nachricht-----
Von: john hamon [mailto:jhamon@xxxxxxxxxxx]
Gesendet: Thursday, January 04, 2001 21:00
An: Bob Fulks; Omega_List
Betreff: RE: AW: back to the well again... parallel trend lines


which, bob is my exact problem and which i carelessly forgot to mention.  i
will map the trend line in and out of arithmetic space to determine my
deltas.  i forgot to mention that.

jh

-----Original Message-----
From: Bob Fulks [mailto:bfulks@xxxxxxxxxxxx]
Sent: Thursday, January 04, 2001 11:48 AM
To: fritz@xxxxxxxx
Cc: omega-list@xxxxxxxxxx
Subject: Re: AW: back to the well again... parallel trend lines


At 12:30 PM -0700 1/4/01, Gary Fritz wrote:

> > Solution: P4 = (x3+x2-x1, y3+y2-y1)
>
>Awww, that's too EASY...  :-)
>
>Good point, Michael.  I got so caught up in trigonometry that I
>didn't think of the obvious approach.
>
>That will produce a line from P3 to P4 at the same angle as the
>original P1-P2 line.  P3 will be one endpoint.  If you wanted the
>parallel line centered around P3, you could just add/subtract 1/2 the
>x2-x1 and y2-y1 values to the P3 location.


Unless you are using a log price scale...