85199 - Survey Sampling

Course Unit Page

SDGs

This teaching activity contributes to the achievement of the Sustainable Development Goals of the UN 2030 Agenda.

Quality education Decent work and economic growth Reduced inequalities Partnerships for the goals

Academic Year 2021/2022

Learning outcomes

By the end of the course the student learns intermediate and advanced topics in finite population inference. The student is able to plan a new survey design and to analyze and summarize the obtained information by a survey with advanced theoretical tools.

Course contents

The topics of the course are illustrated with emphasis to their impact in official statistics.

The source of randomness in sampling from finite populations: design based and model based inference. The predictive approach.

Design based inference for structured populations. Sampling of complex units.

Cluster sampling for clusters of equal and different size. Cluster sampling with varying probability, ratio estimation in cluster sampling. Systematic sampling as a special case of cluster sampling.

Two stage sampling with varying probability, simple random sampling in two stage sampling and primary units with constant size.

Small area inference: design based estimators and Best Linear Unbiased Predictors.

The basic superpopulation models: estimators and their variances.

Area level and unit level models for small area inference.

Approximated bias and variance in design based ratio and regression estimation.

Model assisted design based inference. Generalized regression estimators (GREG) for design-based model-assisted inference. Calibration and balancing.

Multiphase sampling. Consequences of estimating population synthetic quantities in a first sampling phase.

Sampling in space.

Estimation in the presence of non sampling errors.

Readings/Bibliography

Paul S. Levy Stanley Lemeshow (2008) Sampling of Populations: Methods and Applications, 4th Edition, Wiley

 

S.L. Lohr (2010). Sampling: Design and Analysis, Second Edition, Brooks / Cole, Boston.

At the beginning of the course a list of scientific papers is available to students. Each student will choose one of the papers to illustrate at the moment of the exam.

Teaching methods

lectures

Assessment methods

Each student will present the content of the scientific paper chosen during the course. The presentation constitutes 40% of the final mark. Oral questions constitute the other 60% of the mark.

 

The online oral test will be performed via zoom or Teams.

Office hours

See the website of Daniela Cocchi