|
PureBytes Links
Trading Reference Links
|
Hello Dimitris,
sorry for late answer, now I have time for my reply:
> Hans,
> Thank you for your reply and the ref you posted in other messages.
> Here is some notes:
Thank you too Dimitris, I'm sure you are the best ;) to help me and
the group for this.
> 1.
> The formula references FUTURE quotes.
> If you read for ^NDX Oct1 Hurst=0.58, ten days later the value for
> the same date Oct1 WILL NOT BE 0.58 !!
> We have discussed the topic many times in the past, a "HURST"
> research of this list archives will give more than 50 messages.
I was intrested on this "Hurst number" because, as you know, it
represents in some ways the recurrency of a data series, this way it
is possible to select stocks with highest Hurst value.
This code has been posted anywhere, maybe in this Ami Group: I really
do not remember exactly where I copied from.
I can see that there are many corrispondences with theory at
http://www.bearcave.com/misl/misl_tech/wavelets/hurst/
I really cannot judge if the posted code is 100% correct or not.
I also would like to calculate "Hurst exponent" not only for 256
bars, and extend it to for different region sizes, maybe larger or
smaller.
> 1a.
> How do you use this code ?
I use this code in exploration so I can select stocks with highest
hurst exponent.
I lately apply to them several Trading systems, because I
consider "hurst exponent" a quality factor.
> 2.
> If you still want to use it, the code may be substantially shorter .
............
>
> You may continue for the rest of the code in this way.
> It is always interesting to see 10 code lines instead of an A4 page.
> [ If you are interested, I will do the rest of the code for you
some
> other time ]
Thank you for shortening code. As you said, it is always intresting
to see how code can be reduced.
I'm really not able to do it without high risk of damage, thus, I
prefer to have large, but correct code.
My final target is to obtain table with "Hurst Exponent", "Fractal
Dimension", "Lyapunov exponent" and "predictability (day)" columns
exactly as in the as in the "table 1" at
http://www.iqnet.cz/dostal/CHA1.htm
Of course, If you see better use of these "numbers" I'll be happy to
hear about.
> 3.
> The "Chaos Hypertextbook" is interesting. I will translate it in
AFL
> someday. I hope Glenn Elert will be also happy, since the fair use
is
> encouraged [in one way or the other I will ask his
opinion/permission]
I'm looking forward to see the "Chaos Hypertextbook" coded if
possible, starting with these "Lyapunov exponent" and "predictability
(day)".
I will be always grateful to you for your help and your fantastic
attitude to cooperate.
Thanks,
Hans
------------------------ Yahoo! Groups Sponsor --------------------~-->
Make a clean sweep of pop-up ads. Yahoo! Companion Toolbar.
Now with Pop-Up Blocker. Get it for free!
http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/GHeqlB/TM
--------------------------------------------------------------------~->
Check AmiBroker web page at:
http://www.amibroker.com/
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/
|