Multilevel Models and Small-Area Estimation in the Context of Vietnam Living Standards Surveys

• Activity Number: 630
• Type: Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Government Statistics
• Author(s): Dominique Haughton*+ and Phong Nguyen and Irene Hudson and John Boland
• Companies: Bentley University and General Statistics Office, Hanoi and University of South Australia and University of South Australia
• Keywords: Multi-level models ; Small area estimation ; Vietnam ; Living Standards Surveys

Abstract

This talk discusses a methodology to obtain small area estimates in the context of the Vietnam Living Standards Surveys. First we briefly introduce these surveys. Second, we recall main concepts in small area estimation, including the use of auxiliary data, and contrast simple with regression small area models. We discuss random effects in small area regression models, and, in the third part of the talk, present our proposed multilevel model for small area estimation at the commune level in Vietnam, to our knowledge the first such model built with Vietnam living standards data. Our model for estimating the commune-level mean (log of) household expenditure per capita relies on independent variables available both in the 1999 Census and in the VHLSS of 2002 and follows ideas given in work by Moura (1994, 1999); we mention how to measure the accuracy of our model.