Using two-phase sampling scheme, we propose general classes of estimators for finite population variance. These classes differ for the involved sample means and variances of two auxiliary variables. For each class we provide the minimum MSE and show that there exists a chain regression type estimator which reaches this minimum. This estimator, in the widest class, is better than estimators proposed by Singh (1991) and Gupta et al. (1992) , which are optimal in some sub-classes.
Estimation for finite population variance in double sampling.
Tommasi, Chiara;Diana, Giancarlo
2001
Abstract
Using two-phase sampling scheme, we propose general classes of estimators for finite population variance. These classes differ for the involved sample means and variances of two auxiliary variables. For each class we provide the minimum MSE and show that there exists a chain regression type estimator which reaches this minimum. This estimator, in the widest class, is better than estimators proposed by Singh (1991) and Gupta et al. (1992) , which are optimal in some sub-classes.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
WP_2001_22.pdf
accesso aperto
Licenza:
Non specificato
Dimensione
13.85 MB
Formato
Adobe PDF
|
13.85 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.