|
PureBytes Links
Trading Reference Links
|
<FONT face="Comic Sans MS" color=#0000ff
size=4>This is additionally compounded by the fact that different vendors use
different variations for the same formula name.
<FONT face="Comic Sans MS" color=#0000ff
size=4>For example, take stochastics.
<FONT face="Comic Sans MS" color=#0000ff
size=4>
<FONT face="Comic Sans MS" color=#0000ff
size=4>Most software will define this as %K = 100*(Close - Lowest Low of the
last n periods)/(High of the last n periods - Low of the last n
periods).
<FONT face="Comic Sans MS" color=#0000ff
size=4>
<FONT face="Comic Sans MS" color=#0000ff
size=4>Sounds simple enough. Now %D = ema(%K,n) or is it %D = sum of
%K/n divided by the simple average of these terms or do you do
<FONT face="Comic Sans MS"
color=#0000ff size=4>the internal summing and then apply the ema to this result
or perhaps you use Wilder's smoothing instead? All of these definitions
<FONT face="Comic Sans MS"
color=#0000ff size=4>I have seen used as defining stochastic. Also, some
vendors use the terms fast and slow stochastic where slow stochastic is labeled
<FONT face="Comic Sans MS"
color=#0000ff size=4>as %K instead of %D. Now introduce the concept
of rounding and also startup time for time series to converge when using
exponential <FONT
face="Comic Sans MS" color=#0000ff size=4>moving averages and you can see why a
vendor wants verifiable code to implement a solution that can be
backtested. Even <FONT
face="Comic Sans MS" color=#0000ff size=4>backtesting is not that reliable if
indices are being used as they suffer from survivorship bias. Also, if
your stock has been split this <SPAN
class=377220122-30052003>can add
to rounding errors not to mention how volume is treated if your formula(s)
include volume. So even if you can get the original formulas
<FONT face="Comic Sans MS" color=#0000ff
size=4>programmed, then other factors can adversely impact your
result.
<FONT face="Comic Sans MS" color=#0000ff
size=4>
<BLOCKQUOTE
>
<FONT face="Times New Roman"
size=2>-----Original Message-----From: JayTownsend@xxxxxxx
[mailto:JayTownsend@xxxxxxx]Sent: Friday, May 30, 2003 4:48
PMTo: equismetastock@xxxxxxxxxxxxxxxSubject: Re:
[EquisMetaStock Group] Re: Converting Metastock formulas into Excel or
C++<FONT lang=0 face=Arial
size=2 FAMILY="SANSSERIF"><<I need the actual math translation of about
20 indicators. Since I am not a programmer myself, I was (naively?) hoping the
math was easily documented somewhere or there was a way to extract the code
from MetaStock itself. MetaStock customer support doesn't even have many
of the translations! >>It's back to the old bugaboo that many of
these are proprietary even though they are published in works written by the
original authors. Wilder's Parabolic SAR and several other's of his are
in his book, Chakin's are in a book he wrote (if I remember right), Bollinger
Bands are in the writings of Bollinger. All of these sources and many
more are copyrighted and I'm sure that Equis/MetaStock has to pay a royalty in
order to provide them in the software and part of their agreement has to be
that they will not reveal the formulas. When you can determine the
original author, you can go to their books. The King County Library
system (Seattle) had many of them and I'm sure that other libraries do as
well.JayTo
unsubscribe from this group, send an email
to:equismetastock-unsubscribe@xxxxxxxxxxxxxxxYour
use of Yahoo! Groups is subject to the <A
href="">Yahoo! Terms of Service.
Yahoo! Groups Sponsor
To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|