|
PureBytes Links
Trading Reference Links
|
What I
meant was that it is unlikely that any two people doing pairs trading would come
up with the same correlation (82%). There are dozens of formulas and
an infinite number of lookback periods. The reason that I stated
that it wasn't a science is due to my experiences on sites like this
one. If I didn't say that, I would get someone saying that they saw
on www.xyz that IBM has a 81.8% correlation with
GENZ... not 82% as mentioned by me. Then comes the measurement
for how non-correlated those two are as of the close yesterday.
Another lengthy discussion perhaps.
<FONT face=Arial color=#0000ff
size=2>
The
other thing that I surely should have said was that my example WAS NOT A
RECOMMENDATION to make such a trade and that PAST PERFORMANCE IS NOT NECESSARILY
INDICATIVE OF FUTURE RESULTS.
<FONT face=Arial color=#0000ff
size=2>
As to
whether or not pairs trading can work in the future, I don't see it as being any
different from all of the other approaches. There are an (almost)
infinite number of stocks around the globe (I trade on seven exchanges) and an
infinite number of ways to measure correlation and non-correlation.
There's always a way to make money. Of course, you can bias the pair
trading by saying that you would only want to be long IBM and short GENZ for
fundamental, religious or other reasons, ignoring half of the
signals. A gazillion ways to skin this cat.
<FONT face=Arial color=#0000ff
size=2>
I have
one PC dedicated to observing the prices for companies where their stocks trade
in two countries, particularly New Zealand and Australia. The
objective for this PC is to watch for a five-cent difference in price
considering the differences in exchange rates, etc. Buy the stock in
one country, short it another and wait for the five cents to disappear again
before unwinding the deal. This is such a boring way to make money
(two trades a day), that I trust the PC to make all the decisions and execute
the trades. Barely pays for the electricity to power the PC... but
it makes great dinner conversation.
<FONT face=Arial color=#0000ff
size=2>
<FONT face=Arial color=#0000ff
size=2>
<FONT face=Arial color=#0000ff
size=2>
<FONT face=Arial color=#0000ff
size=2>
<BLOCKQUOTE
>
<FONT face="Times New Roman"
size=2>-----Original Message-----From: phsst
[mailto:phsst@xxxxxxxxx]Sent: Sunday, April 20, 2003 12:43
AMTo: amibroker@xxxxxxxxxxxxxxxSubject: [amibroker] Re:
Pairs trading (an example)>It is not a
science. More of an art form.Chuck,The 'art form versus
science' discussion is in just about everypublished work on stock or
futures trading.My take is this:If you can state that there is
about an 82% correlation between IBMand GENZ, and that you used a
'quick scan' to find these two as yournext 'pairs trade'... then we seem
to be talking science rather than'art form', eh?Your
Pal,Phsst--- In amibroker@xxxxxxxxxxxxxxx, "Chuck
Rademacher"<chuck_rademacher@x> wrote:> I just did a quick
scan, looking for some new pairs to trade next week.> > My
method of calculating correlation (there is no real standard)shows
IBM> and GENZ as being about 82% correlated. In the scheme of
things,that means> that they are quite highly correlated.>
> My method of calculating also shows that they have recently
deviatedfrom> their typical correlation enough that I will short
IBM and buy GENZat the> next open.> > If you look at
the two charts, you will probably observe both the> correlation and the
recent deviation from the correlation. It is not a>
science. More of an art form. My pairs trading system shows
thatpair as> the best trade for next week. The system
cannot tell me how longit will> take for the correlation to pop
back in place or, indeed, if it everwill.> It will tell me when to
quit the position, hopefully at a modest profit.> Pairs trading is not
about home runs. It's about slow, consistentequity>
growth with a very low ulcer index. If both of those
companieswent down> the drain or if the whole market has a huge
swing in eitherdirection, I'm> not concerned.> >
CheersSend
BUG REPORTS to bugs@xxxxxxxxxxxxxSend SUGGESTIONS to
suggest@xxxxxxxxxxxxx-----------------------------------------Post
AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A
href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check
group FAQ at: <A
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to the <A
href="">Yahoo! Terms of Service.
Yahoo! Groups Sponsor
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 the Yahoo! Terms of Service.
|