National survey often involve complex sample designs and questionnaire branching. This presents a challenge with imputation of missing data for two reasons. First is the computational intensity of traditional methods, second is imputation methods need to consider data that's Not Missing At Random (NMAR).
New methods namely involving Variational Bayesian Methods offer new approaches to imputation that can address these two ongoing issues in the field of Statistics.
Supervisors
Primary Supervisor: Taylor Winter
Key qualifications and skills
R
Does the project come with funding
No - Student must be self-funded
Final date for receiving applications
Ongoing
How to apply
By email to the primary supervisor
Keywords
Imputation; statistics; Bayesian; missing data
