[R-SIG-Finance] cointegration & reversion to mean

[Paul Teetor]

Stephen,

First, did you get my posting regarding the O-U model and its limitations
when modeling time-to-reversion?

Second, I suggest reading Ernie Chan's book ("Quantitative Trading") if you
haven't already.

Third, I haven't found much research regarding the timing of mean-reversion
trades. It's my impression that academics don't think about that sort of
thing, and professional traders don't publish what they know. That leaves
very little to read.

For my own trading, I want two conditions before taking a mean-reversion
trade. (1) The spread must be at an extreme value, either the top 90th
percentile of its historical range or the bottom 10th percentile of its
range. (2) The spread momentum must be towards the mean, not away.

Those two guidelines don't guarantee a profit, and I'm sure you can find
better ones. But, so far, they've worked for me.

HTH

Paul
 

-----Original Message-----
From: r-sig-finance-bounces at stat.math.ethz.ch
[mailto:r-sig-finance-bounces at stat.math.ethz.ch] On Behalf Of Stephen
Choularton
Sent: Tuesday, October 26, 2010 6:25 PM
To: r-sig-finance at stat.math.ethz.ch
Subject: [R-SIG-Finance] cointegration & reversion to mean

Hi

I am very interested in learning more about cointegration & reversion to
mean when used for spread trading.

I have a working handle on cointegration thanks to

http://quanttrader.info/public/testForCoint.html

but I am getting nowhere with the Ornstein-Uhlenbeck approach to letting me
predict time-to-profit as described in

http://epchan.blogspot.com/2007/01/what-is-your-stop-loss-strategy.html

and it strikes me that there must be some work on how far from the mean the
spread has drifted before it is sensible to take a position but I haven't
found it yet.

If anyone can point me to work on all these subjects I would be most
grateful.
--
Stephen Choularton Ph.D., FIoD

_______________________________________________
R-SIG-Finance at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions
should go.