This paper describes a new procedure to unbiasedly estimate the proportions of t population groups, which at least one is very small and then it can be considered a rare group. This procedure guarantees the privacy protection of the interviewees, as it is based on an extension of the Warner randomized response model. As the estimation regards rare groups, the sampling design considered is the inverse sampling. Some characteristics of the proposed estimators are investigated
A Multi-proportion Randomized Response Model Using the Inverse Sampling
Porro, Francesco
2014-01-01
Abstract
This paper describes a new procedure to unbiasedly estimate the proportions of t population groups, which at least one is very small and then it can be considered a rare group. This procedure guarantees the privacy protection of the interviewees, as it is based on an extension of the Warner randomized response model. As the estimation regards rare groups, the sampling design considered is the inverse sampling. Some characteristics of the proposed estimators are investigatedFile in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.