Title :

ON A GRADUATION PROBLEM INVOLVING BOTH THE HODRICK-PRESCOTT FILTER AND OPTIMAL SPLINE SMOOTHING

Author(s) :

Djehiche, Boualem []
Department of Mathematics [KTH Royal Institute of Technology]
Marcus, Marten [Auteur]
Department of Mathematics [KTH Royal Institute of Technology]
Rahmania, Nadji []
Laboratoire Paul Painlevé - UMR 8524 [LPP]

Journal title :

Far East Journal of Theoretical Statistics

Pages :

1-19

Publisher :

Pushpa Publishing opuse

Publication date :

2011

ISSN :

0972-0863

English keyword(s) :

adaptive estimation
Gaussian process
B-spline function
optimal approximating splines
Hodrick-Prescott filter
Leser filter
noise-to-signal ratio
smoothing
trend extraction
generalized cross validation

HAL domain(s) :

Mathématiques [math]/Statistiques [math.ST]

English abstract : [en]

We suggest an explicit solution to the graduation problem that aims at designing an optimal curve fit which simultaneously behaves as a smoothing B-spline in each time interval (t j-1, t j) and as the HodrickPrescott filter ...
Show more >We suggest an explicit solution to the graduation problem that aims at designing an optimal curve fit which simultaneously behaves as a smoothing B-spline in each time interval (t j-1, t j) and as the HodrickPrescott filter in the knots t j, j = 1,...,n. We also outline its numerical performance and compare it to the existing estimates of the standard Hodrick-Prescott filter.Show less >

Language :

Anglais

Popular science :

Non

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL