Chaoxiong (Michelle) Xia, Ph.D.
Michelle Xia earned her Ph.D. in statistics from the University of British Columbia, her master's degree in actuarial science from the University of Calgary and dual bachelor's degrees —one in insurance, and one in mathematical statistics and probability —from Nankai University.
Besides research, teaching and consulting at NIU, Michelle has over seven years of professional experience as an actuary, a predictive modeler and a statistician in the insurance and medical areas.
Michelle's research is motivated by real-life problems, with current interests covering:
- Machine learning, including deep learning algorithms for health and insurance data.
- Statistical methods for complex data including misrepresentation, misclassification and missing information.
- Predictive analytics, dependence modelling, insurance ratemaking and loss reserving.
- Bias assessment when using imperfect data, including Bayesian methodological work for partially or weakly identified models.
- Computational algorithms including Markov chain Monte Carlo (MCMC) and Expectation Maximization (EM) algorithms.
- Longitudinal and survival methods for complex data.
Gustafson, P., Gilbert, M., Xia, M., Michelow, W., Robert, W., Trussler, T., McGuire, M., Paquette, D., Moore, D., and Gustafson, R. (2013). Impact of statistical adjustment for frequency of venue attendance in a venue-based survey of men who have sex with men. American Journal of Epidemiology, 177, 1157-1164.
Xia, M., and Gustafson, P. (2012). A Bayesian method for estimating prevalence in the presence of a hidden sub-population. Statistics in Medicine, 31, 2386-2398.