In this note an iterated function system (IFS) on the space of distribution functions is built with the aim of proposing a new class of distribution function estimators. One IFS estimator is proposed and its properties are studied in details. Relative efficiency of the estimator, for small and moderate sample sizes, are presented via Monte Carlo analysis. It turns out that the IFS distribution function estimator is, in several cases, more accurate than the celebrated empirical distribution function.
IFS approximations of distribution functions and related optimization problems / S.M. Iacus, D. La Torre. - In: APPLIED SCIENCES. - ISSN 1454-5101. - 6:1(2004), pp. 27-38.
IFS approximations of distribution functions and related optimization problems
S.M. IacusPrimo
;D. La TorreUltimo
2004
Abstract
In this note an iterated function system (IFS) on the space of distribution functions is built with the aim of proposing a new class of distribution function estimators. One IFS estimator is proposed and its properties are studied in details. Relative efficiency of the estimator, for small and moderate sample sizes, are presented via Monte Carlo analysis. It turns out that the IFS distribution function estimator is, in several cases, more accurate than the celebrated empirical distribution function.File | Dimensione | Formato | |
---|---|---|---|
A6-IAC3.PDF
accesso aperto
Tipologia:
Publisher's version/PDF
Dimensione
242.92 kB
Formato
Adobe PDF
|
242.92 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.