A1238
Title: Multivariate generalized regression (GREG) estimation in small area estimation
Authors: Samuel Manda - University of Pretoria (South Africa) [presenting]
Abstract: In survey studies, data from various covariates with known population parameter values are frequently employed to enhance estimation methods. A comprehensive set of auxiliary variables are utilized to explore a multivariate generalized regression (GREG) estimation approach for generating small area estimates across multiple variables of interest. Specifically, the estimation of HIV indicators in South Africa is examined, which includes the proportion of individuals living with HIV who are aware of their status, the proportion of HIV-positive individuals receiving antiretroviral treatment, and the proportion of those on antiretroviral therapy who achieve viral suppression. This technique aims to improve the efficiency and accuracy of population parameter estimates using surveys.
