Yajuan Si

Example image - aligned to the
		right

Assistant Professor

Department of Biostatistics & Medical Informatics and Department of Population Health Sciences

University of Wisconsin-Madison

I was a Postdoctoral Research Scholar working with Andrew Gelman in the Applied Statistics Center and Department of Statistics at Columbia University, NYC. In September 2012, I obtained my Ph.D degree on statistics under the supervision of Jerry Reiter in the Department of Statistical Science at Duke University. My thesis is titled as "Nonparametric Bayesian Methods for Multiple Imputation of Large Scale Incomplete Categorical Data in Panel Studies".

My research lies in cutting-edge methodology development in streams of Bayesian computation, latent variable models, complex survey inference, missing data, causal inference as well as data confidentiality. Currently I am working on hierarchical Bayesian approaches to build a unified framework in sample weighting inferences with particular emphasis in multilevel regression and poststratification, weight smoothing and trimming, deep interactions, borrowing auxiliary information and complex sampling schemes. This approach also provides a path toward the newly-emerging big data explosion problem. In my thesis work, I formalized nonparametric Bayesian multiple imputation framework to deal with panel data of high dimensions and complex dependency structures. My work offers novel strategies for flexible modeling and efficient computation to handle categorical data and non-ignorable attrition.

Selected publications:

Si, Y., Reiter, J. P. and Hillygus, S. (2014), Semi-parametric selection models for potentially non-ignorable attrition in panel studies with refreshment samples, Political Analysis

Si, Y. and Reiter, J.P. (2013), Nonparametric Bayesian multiple imputation for incomplete categorical variables in large-scale assessment surveys, Journal of Educational and Behavioral Statistics, 38, 499 - 521

Deng, Y, Hillygus, S., Reiter, J.P., Si, Y. and Zheng, S. (2013), Handling attrition in longitudinal studies: The case for refreshment samples, Statistical Science, 22, 238 - 256

Si, Y. and Reiter, J.P. (2011), A comparison of posterior simulation and inference by combining rules for multiple imputation, Journal of Statistical Theory and Practice, 5, 335 - 347

Papers under review:

Si, Y., Pillai, N. and Gelman, A. (2014), Bayesian nonparametric weighted sampling inference

Si, Y., Reiter, J. P. and Hillygus, S. (2014), Bayesian latent pattern mixture models for multiple imputation

The following links provide more information about me. Feel free to contact.

Menu:

Contact:

Email: ysi@wisc.edu

Office: Rm 685 WARF Bldg

Address: 610 Walnut St, Madison, WI 53726

Links: