Alex Tsodikov is Professor of Biostatistics. He received his Ph.D. in Applied Mathematics in 1991 from St. Petersburg State Technical University, Russia. Prior to joining University of Michigan, he has been a Postdoctoral Scholar at the Curie Institute (Paris, France), a staff statistician at the University of Leipzig (Germany), Research Assistant/Associate Professor at the University of Utah and Associate Professor/Professor at the University of California, Davis. Dr. Tsodikov's research interests are in various areas of biostatistics and biomathematics, including failure time and survival analysis models, cure models, semiparametric inference, stochastic models, optimal control, inference algorithms based on self-consistency.
BIOSTAT553: Applied Biostatistics
BIOSTAT 875 Advanced Survival analysis
BIOSTAT 560 Statistical Methods for Epidemiology
BIOSTAT 664 Statistics in Cancer Research
BIOSTAT 601 Probability and Distribution Theory
Ph.D., Mathematics/Application of Computing, Mathematical Models and Methods in Natural Sciences, St. Petersburg State Technical University, 1991
M.Sc., Applied Mathematics, St. Petersburg State Technical University, 1988
Research Interests & Projects
My research interests have mainly been evolving around cancer research. My recent methodological interest has centered on the idea of artificial mixture or frailty models and its self-consistency generalization as a tool to derive computationally efficient inference procedures for a wide variety of statistical models. Much of this methodology was initially developed for so-called semiparametric cure models that incorporate improper distributions showing a tail defect. I'm working on a project funded by the National Cancer Institute applying these methods to build a comprehensive model of the dynamics of the national incidence and mortality trends in prostate cancer in the presence of variable utilization of screening. I am also interested in multivariate semiparametric survival models, age-period-cohort models, categorical data analysis and computational approaches to statistical inference such as EM and MM algorithms. I also have a broad ongoing statistical consulting experience in basic and clinical research as well as epidemiology and population sciences.
Search PubMed for publications by Alexander Tsodikov >>
Hu, C., Tsodikov, A. (2013). Semiparametric Regression Analysis for Time-to-Event Marked Endpoints in Cancer Studies Biostatistics
Hu, C., Tsodikov, A. (2014). Joint Modeling Approach for Semicompeting Risks Data with Missing Nonterminal Event Status Lifetime Data Analysis
Ha, J. and Tsodikov, A. (2011). Isotonic Estimation of Survival under a Misattribution of Cause of Death Lifetime Data Analysis, 17
Wang, S., Tsodikov, A. (2010). A Self-consistency Approach to Multinomial Logit Model with Random Effects Journal of Statistical Planning and Inference, 7, 1939-1947.
Tsodikov, A., Chefo, S. (2008). Generalized Self-Consistency: Multinomial logit model and Poisson likelihood Journal of Statistical Planning and Inference, 2380-2397.
Tsodikov, A. and Garibotti, G. (2007). Profile Information Matrix for Nonlinear Transformation Models Lifetime Data Analysis, 1, 139-159.
Tsodikov, A., Szabo, A., and Wegelin, J. (2006). A population model of prostate cancer incidence Statistics in Medicine, 2846-2866.
Tsodikov A. (2003). Semiparametric models: A generalized self-consistency approach. Journal of the Royal Statistical Society, Series B, 65, 759-774.
Tsodikov A. Ibrahim J.G., and Yakovlev A.Y. (2003). Estimating Cure Rates from Survival Data: An Alternative to Two-Component Mixture Models. Journal of the American Statistical Association, 98, 1063-1078.
Tsodikov A.D. (1998). A proportional hazards model taking account of long-term survivors. Biometrics, 54, 1508-1516.
2011 - present: Elected Member, International Statistical Institute
American Statistical Association
International Biometric Society