PureBytes Links
Trading Reference Links
|
Gernot,
it looks like documentation of the alpha factor in EMA formula in AFL.
A two period EMA has an alpha factor of 2/3, a nine period EMA has
alpha of 2/10 or 1/5. If a prior EMA value was 10, and the new data
is 11, and period=9, the new EMA value is (10*4/5)+(11*1/5)
corresponding to (1-alpha)*oldvalue + alpha*newvalue.
--- In amibroker@xxxxxxxxxxxxxxx, "taotra" <ggoll@xxxx> wrote:
> In the T3 afl function in the AB library there is a line of code
> alpha = 2/(n + 1);
> in there, but it doesn't do anything. Anyone know why it's in
there?
>
> Gernot
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Dave Merrill" <dmerrill@xxxx>
> wrote:
> > hi herman, thanks for the feedback.
> >
> > my understanding is that the lack of lag in the jurik stuff makes
> drop-in
> > replacement not necessarily a good idea. depending on how your
> system works,
> > you might need to optimize, maybe even redesign, to make good use
> of them.
> > for example, they say that lag is an important part of the MACD
> mechanism,
> > so replacing all the EMAs in it with JMA is a bad idea; replacing
> only the
> > fast one is better (not sure about the signal one). would you
> think this
> > concern might apply to the tests you did?
> >
> > I'm confused about t3 though. I looked at the AFL version by Dale
> Butkowski,
> > and as a trading system, as written, it didn't appear to perform
> well at
> > all. tons of trades, minimal accuracy, almost no significant
gains
> since
> > '92. is there more than one AFL version? is this the same
> indicator your DLL
> > does?
> >
> > while I've got you, tell me about hbEquity. are there any docs
for
> it? what
> > does it provide that AB's native Equity() function doesn't?
> >
> > thanks,
> >
> > dave
> >
> > > The problem with Jurik is that the product is so "well-guarded"
> that you
> > > can't evaluate it. If you have a specific application you might
> > > try to find
> > > somebody who can run your code on their machine w/jurik stuff,
> to see if
> > > your code performs better using Jurik functions. I managed to
do
> that and
> > > found that in my particular system/application (it may be
> different in
> > > others) it didn't even come close to the simple hbt3a(). BTW,
the
> > > hbt3a() is
> > > a DLL for the T3 published in TASC, nothing secret or magical
> > > about it - it
> > > is also in the AB library in pure afl.
> > >
> > > best regards,
> > > herman
> > >
> >
> > > has anyone tried Jurik's enhanced indicator functions? if so,
> what did you
> > > think?
> > >
> > > they're pricey for sure, relative to AB especially, but do look
> > > interesting.
> > > if they really solve some of the the things they say they do,
> that'd be
> > > great, but that goes for a lot of stuff (:-). I've looked in
the
> list
> > > archive, and didn't find anyone singing their praises, but just
> > > thought I'd
> > > ask. (in fact, that's how I found out about Herman's HBT3A, was
> archive
> > > posts about the Jurik stuff.)
> > >
> > > link is here, if you're interested:
> > > http://www.jurikres.com/catalog/catalog.htm#technical
> > >
> > > thanks,
> > >
> > > Dave Merrill
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for Your HP, Epson, Canon or Lexmark
Printer at Myinks.com. Free s/h on orders $50 or more to the US & Canada. http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/l.m7sD/LIdGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|