Lessons learned from creating a package to create synthetic data
- For whom
- Employees , Students
- When
- 07-12-2021 from 11:30 to 12:30
- Where
- zoom call
- Language
- English
- Organizer
- Jan De Neve
- Contact
- Jan.DeNeve@UGent.be
Lessons learned from creating a package to create synthetic data by dr. Gillian M Raab (Emeritus Professor, Edinburgh Napier University, Part-time Research Fellow, Administrative Data Research Centre - Scotland Edinburgh)
Abstract
Since 2013 we have been working on software to create synthetic data to allow users
of the Scottish Longitudinal study to be supplied with extracts for preliminary analysis
of data that can only usually be accessed from a supervised safe setting. This resulted
in creating the synthpop package for R (www.synthpop.org) that has since been
used to create synthetic data by many other groups world wide. This talk will
give a very brief overview of methods for creating synthetic data and will illustrate
and discuss methods in the package for drawing inferences from synthetic data as
well as our recent work on guidelines for assessing the utility of the synthetic data compared to
the original.
Zoom link
https://ugent-be.zoom.us/j/92105483173?pwd=VmZhamdjb0ZKZGVBcUxILy9FeU41dz09
Meeting ID: 921 0548 3173
Passcode: EL3xwQyd