Professor

Email:

Office: 321 Machray Hall

Telephone: 204-474-6275

Office: 340 Machray Hall

Telephone: 204-474-8172

Personal Webpage

Recent Publications

• Li, X. and Wang, X. (2012). Response adaptive designs with misclassified responses. Comm. Statist. Theory Methods, (in press).
• Li, X. and Wang, X. (2012). Variance-penalized response-adaptive randomization with mismeasurement. J. Statist. Plann. Inference, 142: 2128-2135.
• Wang, X., Tan, Q. and Bickis, M. (2011). Clinical trials with exponential survival times. Comm. Statist. Theory Methods 40, 1510-1524.
• Wang, X. and Wang, Y. (2010). Optimal investment and consumption with stochastic dividends. Appl. Stoch. Models Bus. Ind. 26, 792–808.
• Wang, X. and Yi, Y. (2009). An optimal investment and consumption model with stochastic returns. Appl. Stoch. Models Bus. Ind. 25, 45–55.

→ See more publications

Biography

Xikui Wang received his M.Sc. (Mathematics) and Ph.D. (Statistics) degrees from the University of Saskatchewan, and B.Sc. (Mathematics) degree from Central China (Huazhong) Normal University. He taught at the University of Alberta and Memorial University of Newfoundland before joining the University of Manitoba in 2000. He is a Research Affiliate with the Clinical Institute of Applied Research and Education (CIARE) at the Victoria General Hospital.

Research Programs

1. Response adaptive clinical trials (REACTs) and general biostatistics. A REACT is a data-dependent randomization design which uses information so far accumulated in the trial to sequentially allocate treatments to trial patients. The goal is to improve the efficiency and ethics of the trial without undermining the validity and integrity of the clinical research.
2. Bandit processes (BPs) and Markov decision processes (MDPs), also known as stochastic dynamic programs or controlled Markov processes. A BP is a statistical model of optimal sequential selections from several populations with unknown distributions. The goal is to effectively deal with the competing goals of information gathering (for future benefits) and immediate payoff (for current interests). MDPs study optimal controls of stochastic, dynamic systems. Under the Bayesian approach, a BP can be reformulated as a MDP.
3. Mathematical finance and economics. DPs and MDPs are applied to understand dynamic pricing problems in the face of unknown demand functions and optimal investment and consumption problems with unknown distributions of investment returns.
4. Methods of penalized regression. Penalized regression deals with the methodological and computational challenges of high dimensional data analysis and multicollinearity. The striking issue is the fit-versus-complexity tradeoff. It is a very useful and efficient technique in a variety of applications.

Selected Recent Publications

1. You Liang, Xikui Wang and Yanqing Yi (2013). One-armed bandit processes with a covariate. Annals of the Institute of Statistical Mathematics in press
2. Yi Hua and Xikui Wang (2013). Portfolio selection with a hidden Markov model. Quality Technology and Quantitative Management in press
3. Xuan Li and Xikui Wang (2013). Response adaptive designs with misclassified responses. Communications in Statistics - Theory and Methods 42: 2073-2081.
4. Xuan Li and Xikui Wang (2012). Variance-penalized response-adaptive randomization with mismeasurement. Journal of Statistical Planning and Inference 142: 2128-2135.
5. Xikui Wang, Qier Tan and Mikelis G. Bickis (2011). Clinical trials with exponential survival times. Communications in Statistics - Theory and Methods 40: 1510-1524.
6. David Tolusso and Xikui Wang (2011). Interval estimation for response adaptive clinical trials. Computational Statistics and Data Analysis 55: 725-730.
7. Xikui Wang and Yan Wang (2010). Optimal investment and consumption with stochastic dividends. Applied Stochastic Models in Business and Industry 26: 792-808.
8. Yanqing Yi and Xikui Wang (2009). Response adaptive designs with a variance-penalized criterion. Biometrical Journal 51: 763-773.
9. Xikui Wang and Yanqing Yi (2009). An optimal investment and consumption model with stochastic returns. Applied Stochastic Models in Business and Industry 25:45 – 55.
10. Yanqing Yi and Xikui Wang (2008). Asymptotically Efficient Estimation in Response Adaptive Trials. Journal of Statistical Planning and Inference 138:2899-2905.
11. Xikui Wang (2007). Dynamic Pricing with a Poisson Bandit Model. Sequential Analysis 26:355-365.
12. Xikui Wang, Jeffrey S. Pai and Kevin J. Shand (2007). A Semi-Markov Model of Disease Recurrence in Insured Dogs. Applied Stochastic Models in Business and Industry 23:429-437.
13. Yanqing Yi and Xikui Wang (2007). Statistical Inference of Adaptive Designs with Binary Responses. Communications in Statistics - Theory and Methods 36:1643-1653.
14. Yanqing Yi and Xikui Wang (2007) Goodness-of-fit Test for Response Adaptive Designs. Statistics and Probability Letters 77:1014-1020.

Invited Presentations

1. 2012: Invited talk, Workshop on Probability and Statistics in Finance, University of California, Berkeley, USA
2. 2011: Plenary talk, Fourth International Institute of Statistics and Management Engineering Symposium, Dalian, China
3. 2011: Invited talk, Third International Workshop in Sequential Methodologies, Stanford University, USA
4. 2010: Invited talk, Biostatistics and Epidemiology (Statistical Modelling of Environmental and Health Data), Fredericton, Canada
5. 2009: Plenary talk, International Institute of Applied Statistics Studies in Qingdao, China
6. 2009: Invited talk, Statistical Society of Canada Annual Meeting, Vancouver, Canada
7. 2008: Invited talk, International Conference on Financial Statistics and Financial Econometrics, Chengdu, China
8. 2007: Invited talk, Canadian Mathematics Society Winter Meeting, London, Canada

Current Graduate Students

• You Liang, Ph.D. in progress.
• Yi Eva Hua, Ph.D. in progress.
• Yu Zhang, Ph.D. in progress.
• Han Yu, M.Sc. in progress

Current Research Funding

• Discovery Grant, Natural Sciences and Engineering Research Council (NSERC) of Canada, 2013 – 2018