spring 2012 HEL-8009 Analysis of Incomplete Data and Methods for Imputation - 2 ECTS

  1. Startsida
  2. Emnekatalog
  3. Eldre emner
  4. Analysis of Incomplete Data and Methods for Imputation

Course content

Missing data are ubiquitous in medical research, and raise particular issues as the validity of any analysis depends on inherently untestable assumptions. The aim of this course is to familiarize participants with these issues, and the statistical methods for missing data. Specifically the course will:

  • Explain the issues involved in analysing studies with missing data, review common jargon and argue for a principled, systematic approach.
  • Review a number of ad-hoc approaches for handling missing data and illustrate their limitations.
  • Show how multiple imputation can be used to handle missing data in a principled and practical way.
  • Introduce inverse probability weighting and doubly robust estimation for missing data, and contrast these methods with multiple imputation.

Discuss methods for sensitivity analysis. STATA will be used in the computer practicals.

Objectives of the course

Participants will know the issues raised by missing data and can apply the method of multiple imputation in order to handle missing in a practical, principled way

Language of instruction and examination

English.

Teaching methods

A mixture of lectures and computer practicals will be given.

Assessment

The exam will be a take home test evaluated as passed or failed.

The course will be arranged irregularly and no continuation exams will be given.

Error rendering component

  • About the course
  • Campus: |
  • ECTS: 2
  • Course code: HEL-8009