Sampling and Surveys

Recommended Prerequisites

Basic course in Probability and Statistics.

Teaching Methods

The classes are essentially of expository style and include examples (using real or simulated data) and exercises to apply the material being taught.

There are two types of grading: during the semester or by final exam. During the semester there are two mid-term exams (75 to 100% of the final grade) and a set of homework assignments (until 25% of the final grade) given every two or three other weeks and handed individually. The exercises in the homework assignments are mathematical problems or short computational tasks. The final exam option consists of an exam (75 to 100% of the final grade) and a computational assignment (until 25% of the final grade).

Extensive tutorial time is offered to the students to support the solution of the homework assignments and preparation for the various exams

Learning Outcomes

The main goal is studying the theoretical principles that support the main procedures used in statistical inference for finite populations. The student should understand the limitations and advantages of each sampling design and be able to proof the main properties of the corresponding estimators. The student should also know how to use statistical software to deal with simulated or real data and use Monte-Carlo techniques to compare sampling designs.

This course aims to develop the following skills: ability to calculate, knowledge of mathematical results, ability to formulate and solve problems, design or use of mathematical models to real situations

Work Placement(s)

No

Syllabus

Classical sampling designs: simple random sampling - estimation of totals, means and ratios; stratified sampling; poststratification; cluster sampling; systematic sampling. 

Sampling with unequal probabilities: Hansen-Hurvitz and Horvitz-Thompson estimators;probability proportional to size (PPS) and inclusion probabilities proportional to size (IPPS) sampling.

Other topics in sampling (selection depending on time available): optimality and admissibility; nonresponse; two-phase sampling; complex surveys.

 

Head Lecturer(s)

Carlos Manuel Rebelo Tenreiro da Cruz

Assessment Methods

Assessment 2
A computational assignment : 25.0%
Exam: 75.0%

Assessment 1
The exercises in the homework assignments are mathematical problems or short computational tasks: 25.0%
Frequency: 75.0%

Bibliography

S.L. Lohr, Sampling: Design and Analysis, Duxbury Press, 1999.

Y. Tillé, Théorie des Sondages: Échantillonnage et Estimation en Populations Finies, Dunod, 2001.

A.S. Hedayat, B.K. Sinha, Design and Inference in Finite Population Sampling, Wiley, 1991.

W.G. Cochran, Sampling Techniques, Wiley, 1977.

S.K. Thompson, Sampling, Wiley, 2002.