A abordagem de cadastro duplo (dual frame): estimaÃÃo assistida por modelos lineares com aplicaÃÃo em pesquisas agropecuÃrias

AUTOR(ES)
DATA DE PUBLICAÇÃO

2007

RESUMO

In this thesis, the problem of estimating population totals and means, incorporating auxiliary information at the estimation stage, under a dual frame sampling approach, is considered. A dual frame sampling approach is defined as a survey sample design with two sample frames providing coverage for the same target population. Under this context, several linear model-assisted estimators are proposed for simple random sampling and stratified random sampling designs. Their performance is compared with estimators found in literature and investigated for several dual frame scnenarios by a Monte Carlo simulation study. The proposed estimators are motivated by their potential application to agricultural research

ASSUNTO(S)

dual frame, model-assisted estimador, agricultural research estatistica cadastro duplo, estimador razÃo, pesquisas agropecuÃrias

Documentos Relacionados