Target Participants
Middle and senior level officers in the entire statistical functionaries in a nation will find the course highly rewarding. This includes Bureau of statistics at both state and federal government levels. Departments of Planning, research and Statistics in the entire Ministry, Departments and Agencies (MDAs) both at state and federal government levels.
COURSE RATIONALE
Analytic methods such as small area estimation is useful for producing official statistics. Each small area problem needs to be carefully assessed to ensure that the approach taken and techniques applied suit the particular problem at hand.
Also, if say, small area estimates are to be used more as a guide to indicate areas of unmet demand, then officers saddled with this responsibility must be sufficiently schooled to ensure accuracy. This is the raison d'être for this course.
It is important to train the relevant officials to choose the geographic areas and the key output variables carefully to ensure the results will be fit for the purpose.
This course will help participants to answer questions such as:
COURSE OBJECTIVESParticipants will gain an understanding of the methods for small area estimation, a topic of practical and theoretical interest due to growing demands for reliable small area statistics. Applications will demonstrate the implementation of the methods in practice.The specific objectives are to make participants:
COURSE CONTENT
PART ONESTATISTICAL CONCEPTS
PART TWO• Introduction to Small Area Estimation, Examples of small - area statistics . A perspective on Fay and Herriot ( 1979 ) . Description of some elementary methods and their shortcomings .
PART THREE• Designing surveys for small - area estimation . Spatial similarity . Fine - tuning small - area estimation to a specific policy . Secondary analysis of small - area estimates . The role of simulations and graphics
PART FOUR• Bayesian methods in Small Area Statistics• Robust, semi and non-parametric modeling• Advanced methods in Small Area StatisticsDirect estimation (design based)Direct estimation (model based)Generalized regression (GREG) estimatorBorrowing strength; Indirect estimationsSynthetic methodLink to regressionComposite Estimation• Linear Mixed Models• Small Area (Explicit) Models; Area level model• Small Area (Explicit) Models; Unit level model
PART FIVEData Analysis Using SAS/STAT