-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Devon McCormick" <[EMAIL PROTECTED]> writes:

> Brian - thanks for your help as well. However, perhaps I didn't make it
> clear that I'm looking for the correlations between two separate series: for
> one fund we have monthly return data, for another, only quarterly.

Devon,

Without working out an example, what about using the FFT as an
alternative?  Take the FFT of the input stream, add the appropriate
number of zeros to the middle of the frequency-domain data, and take the
inverse FFT of the padded series.  In your case, you'd need twice as
many zeros as you have quarterly data points; you'd want the total
number of padded points to equal the number of monthly points.

A bit of Googling turned up
http://www.dspguru.com/howto/tech/zeropad.htm as an online reference.
In skimming it, the article looks right.

Don't forget to think about the potential need to low-pass filter and
window your data!

If I were doing this, I might compare the results of this approach with
the structural approach I think you and John were suggesting to see
which helped most.

Bill
- -- 
Bill Harris                      http://facilitatedsystems.com/weblog/
Facilitated Systems                              Everett, WA 98208 USA
http://facilitatedsystems.com/                  phone: +1 425 337-5541
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)
Comment: For more information, see http://www.gnupg.org

iD8DBQFGq8c13J3HaQTDvd8RAurlAJ0TNlYDb5/uFCxdJN0UXAyl+wydMwCeNMhr
zM9/8ZdXEeALr8joNt4CB2E=
=8mlG
-----END PGP SIGNATURE-----

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to