XClose

Institute of Clinical Trials and Methodology

Home
Menu

Using simulation studies to evaluate statistical methods

Next date: TBC, Leicester

Simulation studies are an important tool for statistical research. They help us understand the properties of statistical methods and compare different methods. To perform a meaningful simulation study, careful thought needs to be put into the planning, coding, analysis and interpretation.

Lecturers

Dr Tim Morris, MRC Clinical Trials Unit at UCL, London
Prof Ian White, MRC Clinical Trials Unit at UCL, London
Dr Michael Crowther, University of Leicester and Karolinska Institutet, Stockholm

Course Aims

By the end of the course, participants will be able to:

  • Explain the rationale for simulation
  • Understand the importance of careful planning
  • Have the tools to code and debug simple simulation studies in Stata
  • Know how to analyse simulation studies producing estimates of uncertainty
  • Present methods and results for publication

Target audience

  • Methodological or applied statisticians who need to evaluate the statistical properties of one or more methods
  • PhD students using simulation studies
  • Readers of methodology articles which evaluate methods by simulation

Participants should be familiar with Stata and know, for example, how to generate data, run a regression command and produce simple graphs.

Software

Practicals will be taught using Stata. Attendees will need to bring a laptop with Stata (version 12 or later) installed. If this is likely to be a problem, please get in touch with the course organisers.

Examples

Examples will be taken from the lecturers’ experiences, primarily relating to medical statistics. Principles are the same for simulation studies in other applied areas but the examples may be less relevant.

Registration

Registration for this course will be open in 2018. If you have any queries about the course, please email ictm.cpd@ucl.ac.uk