March 17, 2016, 4:00pm
Location: Department of Statistics and Actuarial Science, University of Waterloo
Room: M3 -3127
Speaker: Jean-François Beaumont, Statistics Canada
Abstract: Although their use has been limited at Statistics Canada so far, Small Area Estimation (SAE) methods exist and offer an alternative to conventional survey methods for addressing local data needs. These methods combine survey data with alternative sources of data, such as administrative data or big data. The goals of this presentation are to provide some basic ideas underlying SAE and to show through an example that it is possible to improve conventional survey estimates, for areas with a small sample size, by complementing survey data with alternative data sources.
This talk is sponsored by the CANSSI CRT Project “Statistical Inference for Complex Surveys with Missing Observations”.
Back to Seminar Series Schedule.