|
PureBytes Links
Trading Reference Links
|
Chuck,
Thanks for the additional information. I get the point about large price drops due
to division spin offs. Back in the tech bubble, I was holding some Bell Canada share
when Bell decided to distribut its shares in Nortel (about 30% of Nortel). The day
the new Nortel shares got credited to my account, my Bell shares dropped by more
than 50%. And you are right, the indicator charts for Bell were useless for many
weeks.
I had an idea while taking my morning shower. Whether the idea is a bright one or I
am just all wet, I do not know yet. But here is the idea.
If Amibroker had one more data field per bar, we could have the following (assuming
the API call to CSI could be explanded):
Date
Open (adjusted for splits, but not for dividends)
High (adjusted for splits, but not for dividends)
Low (adjusted for splits, but not for dividends)
Close (adjusted for splits, but not for dividends)
Volume (actual trading volume, not adjusted for splits)
OI (Actual Close *100, for use in price filters)
NEW (Close adjusted for dividends and for splits)
Alternatively NEW could be set to zero except for the dividend dates when it would
hold the actual dividend payment.
Using the NEW field, one could detect the when a dividend is paided out (by
comparing the daily change in the NEW field to that in the Close field).
What to do when a dividend is detected? Although I am still in a bit of a fog (no
time for breakfast juice), I do not see a simple way to work the dividend change
back into the various indicators and profit and loss calculations.
Man, this dividend stuff is too complicated to tackle before breakfast.
b
--- Chuck Rademacher <chuck_rademacher@xxxxxxxxxx> wrote:
> G'day, b...
>
> Another country heard from on the subject of dividends!
>
> IMO, the BIG dividends could be considered to be rare. However, a dividend
> of 10%, while not BIG, can still affect the P/L and indicator calculations.
> Sometimes a company will float a division or subsidiary, issuing stock in
> the new company and reducing the price of the original stock by the same
> amount. It is not unusual to see 50% falls in price arising from
> situations like this. I could run some stats for you, if you feel it is
> important to know that much about the scale of the problem.
>
> I would hope that TJ might think this subject is important enough to give
> some consideration. In order for that to happen, enough users would have
> to:
>
> a. understand the problem.
> b. believe that it is serious enough that it should be fixed
> c. convey that message to TJ.
>
> If TJ was to address the problem, there still is the issue of how the
> information gets into the database. I hate to bring up CSI again, for fear
> that users here will soon think that I have some sort of financial interest
> in the company. I can assure you that I cannot benefit in any way by people
> subscribing to CSI.
>
> With that (hopefully) out of the way, let me describe how my own trading
> software works. By the way, my software is not for sale, so there is no
> potential conflict in tellying you about it.
>
> My database, thanks to CSI, has a record of every dividend and dividend
> date. This information is available directly from CSI. In fact, with CSI
> you can backadjust your data to include or exclude dividends. Obviously,
> the difference is the amount of the dividend. Backadjusting for dividends,
> however, creates a new problem; prices can go negative when backadjusting
> over several years. My method of getting around this problem is to
> backadjust over a moving window that is large enough to properly calculate
> the indicators being used by the system. Since my database has dividend
> dates and amounts, I am able to do this backadjusting on the fly. Yes, it
> slows things down. No, I don't turn the feature "on" during preliminary
> research of new system(s) and/or parameter(s). I only turn it on in the
> later stages of my research in order to get an accurate picture.
>
> I hope that the above helps people to understand the significance of the
> problem. I don't recall it every being mentioned on this or any other
> Yahoo board. I can assure you, however, that the subject takes up a lot of
> my time and effort.
> -----Original Message-----
> From: b [mailto:b519b@xxxxxxxxx]
> Sent: Tuesday, January 06, 2004 1:59 AM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: RE: [amibroker] dividend (more on the subject)
>
>
> Chuck,
>
> I am convinced. Dividends are bad news. And with recent US tax changes
> they will not
> be going away anytime soon.
>
> Are the big dividend payouts rare enough that they can be ignored in
> practice?
>
> I hope the answer is Yes, because I can not see any easy way to compensate
> (other
> than manually checking all trade candidates for recent dividends - too
> much tedious
> work for my liking.)
>
> b
>
> --- Chuck Rademacher <chuck_rademacher@xxxxxxxxxx> wrote:
> > Another reason why I say that dividends are the BIGGEST PROBLEM with
> data is
> > that not only did $11 disappear in one day from the stock in question
> > (causing a loss), it also severely distorts ALL technical indicators.
> Such
> > a move could look like a SERIOUS BREAKOUT when, in fact, it's nothing
> but a
> > dividend.
> > -----Original Message-----
> > From: indiana0352 [mailto:cs_winn@xxxxxxxxxxx]
> > Sent: Monday, January 05, 2004 9:34 PM
> > To: amibroker@xxxxxxxxxxxxxxx
> > Subject: [amibroker] dividend
> >
> >
> > Does anyone know how I can compensate for large dividend payments
> > made by companies in the stock prices in AB??
> >
> > eh NCRX which paid out a $10.92 cash dividend on Dec 2, shows in AB
> > as dropping significantly that day when really it didn't.
> >
> > The only way I could firgure it out would be to do a reverse
> > division and calculate a split value which I enter. Is there an
> > easier way?
> >
> > TIA,
> > Chris
>
__________________________________
Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
http://hotjobs.sweepstakes.yahoo.com/signingbonus
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
Yahoo! Groups Links
To visit your group on the web, go to:
http://groups.yahoo.com/group/amibroker/
To unsubscribe from this group, send an email to:
amibroker-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|