BAYESIAN ANALYSIS OF KIM AND WARDE RANDOMIZED RESPONSE TECHNIQUE USING ALTERNATIVE PRIORS (Review Completed - Accepted)
In this paper, we developed the Bayesian estimators of the population proportion of a stigmatized attribute using Kumaraswamy and Generalised Beta prior distributions when data were obtained through the Randomized Response Technique (RRT) proposed by Kim and Warde [15]. We validated our newly developed Bayesian estimators for a wide range of the designed values of the population proportion at varying sample sizes. It was observed that our newly developed Bayesian estimators performed significantly better than the Bayesian estimator developed by Hussain and Shabbir [12] for relatively small as well as moderate sample sizes. However, the reverse was the case for very large sample sizes.
Keywords: Absolute Bias., Alternative priors, Bayesian estimation, Mean Square Error (MSE), Stigmatized attribute