Zhiguo Li
Research Fellow
Quantitative Methodology Program
Institute for Social Research
University of Michigan
E-mail: zhiguo@umich.edu
Education
PhD in biostatistics, University of Michigan, 2008
Research Interests
- dynamic treatment regimes
- survival analysis
- semiparametric models
- missing data
Publications
- Comparing an experimental agent to a standard agent: relative merits of a one-arm and a randomized two-arm phase II design, Taylor, J.M., Braun T.M. and Li Z., Clinicla Trials, 2006,3:335-348.
- Analysis of bianry responses with ordered covariates and missing data, Taylor J.M., Wang L. and Li Z., Statistics in Medicine, 2007, 3443-3458.
- Weighted likelihood method for grouped survival data in case-cohort studies, with application to HIV vaccine trials, Li Z., Gilbert P. and Nan B., Biometrics, 2008, in press
- Missing covariates in Cox regression with current status data, Li Z. and Nan B., submitted
- Construction of confidence intervals and regions for ordered binomial probabilities, Li, Z., Taylor, J.M. and Bin N., in revision
- Development of anxiety after antidepressant use among depressed veterans, Li Z., Pfeiffer N.P., Hoggatt J.K., Zivin K., Downing K., Ganoczy D., and Valenstein M., submitted
- Sample size calculation for comparing two-stage treatment strategies with censored data, Li Z. and Murphy S.A., submitted
Current Research
I'm currently working with Dr. Susan Murphy in survival analysis with multi-stage randomization. The problem we are particularly interested in is to come up with a convinient way of determining the sample size needed for a clinical trial with two-stage randomization and survival endpoints. Click here for a poster presentation on this work at ENAR 2009 and the SCT 2009 Annual Meeting, and here for a paper on this work. I'm also collaborating with researchers in psychiatry and epidemiology in a survival analysis project using data from the Veteran Administration. The main interest is the causal inference of antidepressant drugs and suicide. Click here for a poster presentation of this work at the Sliverman Conference, 2009.
Some personal stuff