Bayesian Bandwidth Estimation based on Ranked Set Sampling

نویسندگان

1 Department of Statistics, Allameh Tabataba'i University, Tehran, Iran

2 Department of Statistics, Allameh Tabataba'i University, Tehran, Iran.

doi
10.22034/jirss.2024.2004038.1024
چکیده

In the estimation of a probability density function (PDF) by kernel method, two inherent problems are the choice of sampling methods and the selection of a bandwidth. In this article, we use the balanced and unbalanced ranked set sampling (RSS) methods and Bayesian bandwidth to estimate a PDF by kernel method. To compare our method with existing methods, we use an extensive simulation study to compare the RSS with simple random sampling (SRS) PDF estimator and also Bayesian bandwidth with other existing bandwidths. As an application, we use the household expenses and income data of the Statistical Center of Iran in 2021 to estimate the PDF of the total expenses of the households in Tehran province.