Estimation of population total using model-based approach: a case of HIV/AIDS in Nakuru central district, Kenya

Abstract

: In this study we have explored an estimator for finite population total under the famous prediction approach. This approach has been compared with design-based approach using simple random sampling and stratified random sampling techniques. It is shown that the estimators under model based approach give better estimates than the estimators under design based approach both when using simple random sampling (s.r.s) and stratified random sampling. The relative absolute error from both approaches is computed and has been shown to be superior under the super population model than the design based approach. This approach is then applied to predict the total number of people living with HIV/AIDS in Nakuru Central district