Displaying 1 result from an estimated 1 matches for "fracfiff".
Did you mean:
fracdiff
2001 Mar 31
0
confused about range of 'd' in fracdiff package
Dear all,
I want to assess the question whether several time series of parties'
respective popularities are fractionally integrated. The "fracdiff"
package seems to be an obvious choice. What confuses me is that the
'd' parameter estimated by fracfiff seems to be bound to a range from
0 to 0.5. From what I have read I would assume it should be allowed to
vary between 0 and 1.
Any help would be very much appreciated.
Kai
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuw...