The effect of the practice of rounding non-integer weights in bootstrapping complex samples from finite populations is investigated by means of an extended simulation study. The extent to which rounding can interfere with basic Bootstrap principles as well as with the formal properties of the final Bootstrap estimates are evaluated. Indications and recommendations on application of this practice are discussed.
Rounding Non-integer Weights in Bootstrapping Non-iid Samples: actual problem or harmless practice? / F. Andreis, F. Mecatti - In: Advances in Complex Data Modeling and Computational Methods in Statistics / [a cura di] A. M. Paganoni, P. S. Secchi. - [s.l] : Springer, 2015. - ISBN 978-3-319-11149-0. (( convegno S.Co.2013: Complex data modeling and computationally intensive methods for estimation and prediction tenutosi a Milano nel 2013 [10.1007/978-3-319-11149-0_2].
Rounding Non-integer Weights in Bootstrapping Non-iid Samples: actual problem or harmless practice?
F. Andreis;
2015
Abstract
The effect of the practice of rounding non-integer weights in bootstrapping complex samples from finite populations is investigated by means of an extended simulation study. The extent to which rounding can interfere with basic Bootstrap principles as well as with the formal properties of the final Bootstrap estimates are evaluated. Indications and recommendations on application of this practice are discussed.
| File | Dimensione | Formato | |
|---|---|---|---|
|
329415_1_En_2.pdf
accesso riservato
Descrizione: Pre-publication proof
Tipologia:
Altro
Dimensione
453.07 kB
Formato
Adobe PDF
|
453.07 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
