[R-sig-finance] Baxter-King filtering?

[Andrew Piskorski]

On Fri, Nov 25, 2005 at 07:03:40AM +0530, Ajay Narottam Shah wrote:
> A few weeks ago, Gabor showed this code for Hodrick Prescott
> filtering:

> The literature seems to think that Baxter-King works better. The idea

This paper:

  "A Frequency Selective Filter for Short-Length Time Series"
  by Alessandra Iacobucci, May 2004
  http://www.ofce.sciences-po.fr/pdf/dtravail/WP2004-05.pdf

compares the Hodrick-Prescott, Baxter-King, and Christiano-Fitsgerald
filters to his Hamming-windowed filter, and says that the
Hamming-windowed is best.

-- 
Andrew Piskorski <atp at piskorski.com>
http://www.piskorski.com/