Linkletter, Crystal

Crystal D Linkletter

Adjunct Assistant Professor of Biostatistics

Overview

Dr. Linkletter arrived at Brown after completing her doctorate in Statistics from Simon Fraser University in Burnaby, British Columbia. She is now a faculty member in the Biostatistics Section of the Department of Community Health and in the Center for Statistical Sciences. Dr. Linkletter also comes to Brown as an affiliate of the S4 Initiative.

Her research interests include statistical modeling of social networks, Bayesian hierarchical modeling, spatial statistics, statistical analysis of computer experiments, experimental design, and industrial statistics. Many of these interests arose from various research assistantships at the Los Alamos National Laboratory. During her doctoral studies, Dr. Linkletter received funding from the National Science and Engineering Research Council of Canada and the National Program on Complex Data Structures.

Currently Dr. Linkletter teaches courses in Bayesian Statistical Methods and Spatial Data Analysis, which are newly offered by the Biostatistics section in Community Health. She also teaches a course in Applied Regression Analysis.

Dr. Linkletter earned a BScH in Mathematics and Statistics from Acadia University in Wolfville, NS and an MS in Statistics from Simon Fraser University.

Brown Affiliations

Biostatistics

Research Areas

Research

Research Overview

Dr. Linkletter develops statistical methods for social network analysis, with emphasis on predicting network connections and understanding how contact structure impacts processes evolving on the network. Other projects include design and analysis of experiments, and statistical modeling of computer experiments.

Research Statement

Dr. Linkletter's research interests today broadly include social network modeling, Bayesian hierarchical modeling, design and analysis of computer experiments, and industrial statistics. Key aspects of her work are developing statistical methods to help predict social network connections, and understanding the role "contact networks" play in the facilitation of dynamic processes (such as the spread of infectious diseases).

Scholarly Work

Fiecas, M., Ombao, H., Linkletter, C., Thompson, W. and Sanes, J. (2009). Functional connectivity: Shrinkage estimation and randomization test. Submitted to NeuroImage.

Dizon, D., Schutzer, M., Politi, M., Linkletter, C., Miller, S., and Clark, M. (2009). Advance care planning decisions of women with cancer: provider recognition and stability of choices. To appear in J Psychosocial Oncology.

Linkletter, C., Ranjan, P., Lin C., Bingham, D., Brenneman, W., Lockhart, R. and Loughin, T. (2009). Compliance testing for random effects models with joint acceptance criterion. Tentatively accepted with revisions to Technometrics.

Guillas, S., Rougier, J., Maute, A., Richmond, A.D. and Linkletter, C.D. (2009). Bayesian calibration of the thermosphere-ionosphere electrodynamics general circulation model (TIE-GCM). Submitted to Geoscientific Model Development.

Linkletter, C.D. and Sitter, R.R. (2009) Latent socio-spatial process model for social network analysis. Submitted.

Linkletter, C.D. (2008). Social network analysis: practical and statistical challenges. Health Services and Outcomes Research Methodology, 8, 270-272.

Linkletter, C.D. and Sitter, R.R. (2007). Predicting natural gas production in Texas: an application of nonparametric reporting lag distribution estimation. Journal of Official Statistics, 23, 239-251.

Linkletter, C., Bingham, D., Hengartner, N., Higdon, D. and Ye, K.Q. (2006). Variable selection for Gaussian process models in computer experiments. Technometrics, 48, 478-490.

Background

Education and Training

Year	Degree	Institution
2007	PhD	Simon Fraser University

Honors and Awards

Jack Youden Prize (Chemical Division of the American Society for Quality) for the best expository paper in Technometrics, 2006:

Linkletter, C., Bingham, D., Hengartner, N., Higdon, D. and Ye, K.Q. (2006). Variable selection for Gaussian process models in computer experiments. Technometrics , 48 , 478-490.

Researchers@Brown