Hi All
I would be interested in hearing from anyone who has implemented, or knows
about the existence of, S code for spline smoothing in the presence of
autocorrelation. I'm particularly interested in the approaches of
Diggle and Hutchinson 1989 On spline smoothing with autocorrelated errors.
Austral. J. Statist. 31:166-82
Donnelly, Laird, and Ware 1995 Prediction and creation of smooth curves for
Temporally Correlated Longitudinal Data. JASA 90:984-89
D&H generalise the cross-validation criteria to incorporate simultaneous
estimation of smoothing parameter and autocorrelation structure. D,L&W lend
from kriging to model temporal correlation as a special case of spatial
correlation.
Any help/pointers would be most appreciated. Apologies to those people who
sort their mail by date - strange as this may seem, changing my system date
is beyond my control at the moment.
John McKinlay
NT4; V3.3
Real date: 25 May 98
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|