Research at FAST

95 Department of Applied Mathematics Department of Applied Mathematics Qualification PhD (HKU) Postdoc (NIH) ORCID ID 0000-0003-3410-445X Representative Publications • Shen, J., Yu, H., Yang, J., & Liu, C *. ( 2019 ). Semiparametric Bayesian analysis for longitudinal mixed effects models with non-normal AR (1) errors. Stat. Comp., 29(3), 571-583 • Zhang, Z., Ma, S., Shen, C., & Liu, C . ( 2019 ). Estimating Mann–Whitney‐type Causal Effects. Intern. Stat. Rev., 87(3), 514-530 • Lee, J. W., Liu, C ., Chan, J. C., & Lai, J. S. ( 2014 ). Predictors of success in selective laser trabeculoplasty for chinese open-angle glaucoma. J. Glaucoma, 23(5), 321-325 • Liu, A., Liu, C ., Zhang, Z., & Albert, P. S. ( 2012 ). Optimality of group testing in the presence of misclassification. Biometrika, 99(1), 245-251 • Liu, C ., Liu, A., & Halabi, S. ( 2011 ). A min–max combination of biomarkers to improve diagnostic accuracy. Stat. Med., 30(16), 2005-2014 Awards and Achievements • Honorary Assistant Professor (11/ 2013 -06/ 2016 ), Department of Ophthalmology, University of Hong Kong • Intramural Research Training Award Postdoctoral Fellow (10/ 2008 -07/ 2010 ), National Institutes of Health, USA Professional Services • Guest Editor for the special issue of Biostatistics, Computational Statistics and Data Analysis • Associate Editor for the journal Computational Statistics and Data Analysis • Associate Editor for the journal Statistics and its interface Dr LIU Catherine Chunling Associate Professor Research Overview Statistical theory and methodology: Bayesian Statistics; Incomplete data analysis: censoring, missing, and limit of detection; Functional/ Longitudinal/ Multivariate /high-dimensional/image data analysis; Non- and Semi-parametric modelling and inference; Sequential statistics. Empirical studies: diabetes mellitus, glaucoma in ophthalmology, Alzheimer Disease, hydrology, environmental epidemiology. The PI’s ongoing project focuses on detecting Single Nucleotide Polymorphisms (SNPs) level genotype effects on the phenotype response in longitudinal Genome-Wide Association Studies (GWAS). For non-Gaussian response situations, the multiple level test is challenging and is hard to reach GWAS significant level threshold. Functional data analytics shed light on providing powerful test procedures that is GWAS significant and still achieve the statistical optimality. Some generic results can be detected for patients of Alzheimer Disease at mild cognitive impairment stage and ADNI1 cohort separately by two sets of functional ANOVA test procedures.