Small‐area Estimation with Applications to Agriculture

This chapter provides a brief account of small-area estimation in the context of agricultural surveys. Preventive measures at the design stage may significantly reduce the need for indirect estimates, but for some small areas sample sizes may not be large enough to provide adequate precision using direct estimates even after implementing such measures. The chapter turns to model-based methods based on small-area linking models involving random small-area effects. Such models may be broadly classified into two types: (a) area-level models and (b) unit-level models. The chapter emphasizes the need to formulate an overall programme that covers issues related to sample design and data development, organization and dissemination, in addition to those pertaining to methods of estimation for small areas. Controlled Vocabulary Terms estimation; surveys