Niko Kaciroti

nicola's picture
Token Available
University: 
University of Michigan - Ann Arbor
Unit: 
UMOR Units
Department: 
Center for Human Growth & Development (CHGD)
Title: 
Research Scientist, CHGD and Associate Research Scientist, Biostatistics, SPH
Phone: 
734-764-2443
Short bio: 

Ph.D., Biostatistics, 2002

Research summary: 

 

Niko Kaciroti is a Research Scientist at the Center for Human Growth and Development and the Department of Biostatistics. Dr. Kaciroti's research interest is in analyzing longitudinal and repeated measures data with missing values. He is currently working on developing and using Bayesian models for analyzing longitudinal outcomes with nonignorable missing data. Another area of research is using complex and dynamic models within a hierarchical Bayesian framework to analyze cortisol data. His applied research is focused in the application of statistics in an interdisciplinary setting related to medical, social science and public health fields. This includes, application of advance statistical methods to longitudinal data using linear and nonlinear mixed models, survival analysis, structural equation modeling, and path analysis.  

Recent publications: 

 

Kaciroti N, Raghunathan TE. Bayesian sensitivity analysis for incomplete data: bridging pattern-mixture and selection models for exponential family. Statistics in Medicine. 2014;33(27):4841-57. 

Kaciroti N, Raghunathan TE, Taylor J, Julius S.   (2012) A Bayesian model for discrete time-to-event data with informative censoring.  Biostatistics.2012; 13(2): 341-354. 

Kaciroti N, Schork MA, Raghunathan TE, Julius S. (2009). A Bayesian Sensitivity Model for Intention-to-Treat Analysis of Binary Outcomes with Dropouts. Statistics in Medicine28(4), 572-5.

Kaciroti N, Raghunathan TE, Schork MA, Clark NM. A (2008). Bayesian model for longitudinal count data with non-ignorable dropout. Journal of the Royal Statistical Society C: Applied Statistics.57, 521-534.

Kaciroti N, Raghunathan TE, Schork MA, Clark NM, Gong M. (2006). A Bayesian Approach for Clustered Longitudinal Ordinal Outcome with Nonignorable Missing Data: Evaluation of an Asthma Education Program. Journal of American Stat. Assoc.101, 435-446.

Keywords: 
missing data,
nonlinear and dynamic models,
path analysis,
clinical trials,
Bayesian analysis,
longitudinal data