The Generalized Logistic Regression Estimator in a Finite Population Sampling without Replacement Setting with Randomized Response El estimador de regresi'{o}n log'{i}stica generalizado en un muestreo sin reemplazo con respuesta aleatorizada en poblaciones finitas

The randomized response technique (RR), introduced by Warner (1965) was designed to avoid non-answers to questions about sensitive issues and protect the privacy of the interviewee. Some other randomized response techniques have been developed as the Mortons technique which was developed based on a...

Full description

Bibliographic Details
Main Authors: VÍCTOR HUGO SOBERANIS-CRUZ, VÍCTOR MIRANDA-SOBERANIS
Format: Article
Language:English
Published: Universidad Nacional de Colombia 2011-01-01
Series:Revista Colombiana de Estadística
Subjects:
Online Access:http://www.scielo.org.co/scielo.php?script=sci_arttext&pid=S0120-17512011000300004
Description
Summary:The randomized response technique (RR), introduced by Warner (1965) was designed to avoid non-answers to questions about sensitive issues and protect the privacy of the interviewee. Some other randomized response techniques have been developed as the Mortons technique which was developed based on a finite population sampling without replacement. In this paper we are presenting an estimation of the population (total of individuals N) based on Mortons technique assisted for a logistic regression model and considering a specific sensitive characteristic A, with an auxiliary variable associated to the sensitive variable. Analyses were conducted assuming finite population sampling and based on the p-estimators theory through a model assisted estimator. In addition, we propose an estimator of the variance of the estimator, as well as the results of simulations showing that the model assisted estimator of the variance decreases compared with an estimator which depends of the sampling design.<br>La t{e}cnica de respuesta aleatorizada (RA) introducida por Warner (1965), fue diseñada para disminuir la no-respuesta sobre aspectos sensibles y para proteger la confidencialidad del entrevistado en muestreos con reemplazo. Otras t{e}cnicas RA para muestreos sin reemplazo en poblaciones finitas, como la de Morton, han sido desarrolladas y comparadas. En este trabajo se exponen los resultados de la estimaci{o}n del total de individuos de una poblaci{o}n finita con la t{e}cnica de Morton, considerando una caracter{i}stica espec{i}fica sensitiva A en un muestreo sin reemplazo y asistido por un modelo de regresi{o}n log{i}stica, con una variable auxiliar asociada a la variable sensitiva. Se desarrolla en el contexto de poblaciones finitas y en el marco de la teor{i}a de los estimadores-&pi;, a trav{e}s de un estimador asistido por el modelo. Asimismo, se propone un estimador para la varianza del estimador y se muestra, vía simulaci{o}n, que este estimador para la varianza disminuye, comparado con otro estimador que depende {u}nicamente del diseño de muestreo.
ISSN:0120-1751