|
PureBytes Links
Trading Reference Links
|
Hi,
I think that Dimitris pointed out one important
thing:
any drawing tool that uses "angle" in its
composition:
for example "trendline by angle", "Gann 45%
line",
"Gann grid" are somewhat misleading because
changing one axis scale changes all angles :-)
That's why I am not a big fan of implementing
these
in AmiBroker.
Best regards,Tomasz
Janeczko------------------------------------------------AmiBroker -the
comprehensive share manager<A
href="">http://www.amibroker.com
<BLOCKQUOTE
>
----- Original Message -----
<DIV
>From:
Dimitris
Tsokakis
To: <A title=amibroker@xxxxxxxxxxxxx
href="">amibroker@xxxxxxxxxxxxxxx
Sent: Sunday, August 12, 2001 2:49
PM
Subject: [amibroker] Angles
In att gif A is a local minimum and B isa local
maximum.
The two graphs are EXACTLY the same "Price"
graph
for the selected stock.
The angle at the above graph looks like a fine 45
deg.
The same angle in below graph looks like30
deg.
What is the real angle and how will be the same
for
both graphs.
Is there any "standard" ratio that we should
follow with
the accuracy of Euclidean Geometry in angles
design?
Is there any "standard" goniometer superimposed
on
the "standard" graph, in order to trace the
"accurate"
45 deg line?
Of course not.
Angles, as two dimensional objects change their
visual
shape according to the climax of each
axis.
And here is a great
misunderstanding.
In Analytic Geometry, the x´x and y´y axis
represent the
same set of Real numbers, so they have the same
unit 1
and the line from (0,0) to (3,3) will ALWAYS have
angle 45 deg.
But, here is not the same.
x´x axis is time and for a daily graph the unit
is 1 day.
y´y axis is price and the unit is 1 USD or 1 EURO
or 1
unit of Dow Jones scale etc.
A square in Analytic Geometry means 4 right
angles and
all sides equal (RELATED TO THE SAME
UNIT)
A square here has two sides equal to 35 USD and
two sides
equal to 25 DAYS.
To ask 4 sides equal is nonsense.
But, if you want to forget the Euclideandream
and if you want
the deal with REAL properties, there is a
solution.
In att. gif we have
tA=282 days
tB=302 days
tB-tA=40 days
highB=6.56 euros
lowA=4.56 euros
gain=2 euros
The ratio R=gain/(tB-tA)=0.05 will be independent
of any shape
of your graph, you may call it "tangent"of the
hypothetical "angle"
and use it as a measure for further correlations
for the same stock.
If in the next uptrend the stock gains 1.5 euro
in 30 days, then this
ratio will be 0.05 and the "angles" willlook the
same on any graph
at any scale.
Just keep in mind that this "angle" willnever
have a relation with
Euclidean angles and any sin, cosin or atan
function will NOT give
the expected results.
No matter if we like it or not, Euclidean
Geometry presupposes
common measure for both axis.
Dimitris Tsokakis
Your use of Yahoo! Groups is subject to the <A
href="">Yahoo! Terms of Service.
|