Statistical Sampling Theory (Course code 0413)

Purpose

To familiarize participants with sampling methods and their application.

Benefits to participants

Participants will gain an understanding of the Statistical Sampling Theory.

Target population

Employees who have to develop and implement complex sampling plans in their work and wish to get an advanced knowledge of sampling theory.

Course outline

  • Classical statistics vs. finite population inference
  • Sampling designs, inclusion probabilities
  • Horvitz-Thompson estimator and its properties
  • Monte Carlo simulations
  • Stratification, sample allocation, construction of strata, introduction to GSAM
  • Inclusion probability proportional to size sampling, high entropy sampling designs
  • Balanced sampling, the Cube algorithm
  • Estimation of complex parameters, Taylor linearization procedures
  • Ratio estimator, post-stratified estimator, raking ratio estimator, difference estimator, generalized regression estimator
  • Multistage sampling
  • Two-phase sampling
  • Model-based inference

Prerequisite

Advanced knowledge of mathematical statistics and basic knowledge in sampling theory.

Delivery type: Virtual instructor-led

Duration: 21 days (2hrs per sessions for 10 weeks)

Contact:
If you have questions or to register to the course, contact us at statcan.msmdsstatstraining-msmsdformationstats.statcan@statcan.gc.ca

Date modified: