Vai ai contenuti. | Spostati sulla navigazione | Spostati sulla ricerca | Vai al menu | Contatti | Accessibilità

| Crea un account

Di Fonzo, Tommaso - Marini, Marco (2011) A Newton's Method for Benchmarking Time Series according to a Growth Rates Preservation Principle. [Working Paper] WORKING PAPER SERIES, 7/2011 . , PADOVA (Inedito)

Full text disponibile come:

[img]
Anteprima
Documento PDF
1001Kb

Abstract (inglese)

We present a new technique for temporally benchmarking a time series according to the Growth Rates Preservation (GRP) principle by Causey and Trager (1981). This procedure basically looks for the solution to a non linear program, according to which f(x), a smooth, non-convex function of the unknown values of the target time series xt, t = 1, . . . , n, has to be minimized subject to linear equality constraints which link the more frequent series xt to a given, less frequent benchmark series bT, T = 1, . . . ,m. We develop a Newton's method with Hessian modification applied to a suitably reducedunconstrained problem. This method exploits the analytic Hessian of the GRPobjective function, making full use of all the derivative information at disposal.
We show that the proposed technique is easy to implement, computationally robust and efficient, all features which make it a plausible competitor of other benchmarking procedures (Denton, 1971; Dagum and Cholette, 2006) also in a data-production process involving a considerable amount of series.


Statistiche Download - Aggiungi a RefWorks
Tipo di EPrint:Working Paper
Anno di Pubblicazione:Giugno 2011
Parole chiave (italiano / inglese):Benchmarking, Movement preservation, Linearly equality constrained non-linear optimization, Newton's method. JEL classification codes: C22, C61, C82.
Settori scientifico-disciplinari MIUR:Area 13 - Scienze economiche e statistiche > SECS-S/01 Statistica
Struttura di riferimento:Dipartimenti > Dipartimento di Scienze Statistiche
Codice ID:7190
Depositato il:18 Nov 2014 14:58
Simple Metadata
Full Metadata
EndNote Format

Download statistics

Solo per lo Staff dell Archivio: Modifica questo record