B015129 - SAMPLING THEORY

Italiano/Inglese

The course deals with the theory and methods for selecting samples from finite populations and making inferences on the main population parameters.
It copes with the problem of parameters estimation in presence of non-sampling errors, mainly those due to non-response.
Moreover, it deals with the problem related to the estimation of the variance/uncertainty of the estimators in complex sample designs.

The teacher will suggest several readings during the course.

The aim of the course is to introduce students to the methods and tools needed to design and run a sample survey as well as to the methods and tools needed in order to use the survey data for producing reliable estimates.

After completing the course, the students should be able to:

- take the key decisions relating to the choice of a sampling design

- be responsible for conducting a statistical survey

- evaluate alternatives available for estimating population parameters

- produce the results of a survey taking into account the non-sampling errors.

Knowledge of basic elements of probability theory and statistical inference.

Lectures, exercises and computer sessions.

Written test with questions about the theory and exercises, plus an additional oral examination.

- Introduction to finite population sampling

- Probability samples; sample design; inclusion probability; sample membership indicators; main sampling designs; estimators for the main population parameters.

- The role of auxiliary variables in sampling design and estimation;

- Design based estimation vs model based estimation; design assisted estimation.

- Estimation in complex surveys and empirical variance estimation.

- Estimation in the presence of non-response and frame imperfections.