- Professional overview
Gary Yu’s dissertation focused on creating a new statistical technique for clustering individuals based on their patterns of responses (e.g. on a questionnaire of drug items). He extended the finite mixture model to allow for the number of repeated measures to be incorporated and to contribute to the clustering of individuals. The dimension of the repeated measures can be summarized into a count of responses and can be assumed to follow a truncated Poisson distribution and this information can be included into what is called a dimension informative finite mixture model (DIMM) [NIH/NHLBI R01HL111195]. This model was originally developed and applied to continuous physical activity data and it has been applied to binary drug items among men who have sex with men (MSM) in the US and among male sex workers (MSW) in Vietnam.
DrPH, 2014, Biostatistics, Columbia UniversityMPH, 2006, Epidemiology, Boston UniversityBS, 2004, Bioengineering, University of California, Berkeley
- Honors and awards
Bernard Challenor Spirit Prize, Mailman School of Public Health, Columbia University (2014)
No publications could be found that match your search criteria.