My main research areas are firmly within the realm of Bayesian statistics. I’m generally interested in models for dependent data, especially time series, spatial, and spatial temporal models, with a focus on applications to official statistics. Often the application is to find better ways to extract more information out of existing data products published by official statistical agencies. I also have an interest in Markov chain Monte Carlo methods, especially with application to the sort of models I’m interested in. But my typical MCMC strategy (and advice) is to use Stan whenever possible.
-  Adaptively tuned particle swarm optimization with application to spatial design, STAT (Matthew Simpson, Christopher K. Wikle, and Scott H. Holan)
-  Interweaving Markov Chain Monte Carlo Strategies for Efficient Estimation of Dynamic Linear Models, Journal of Computational and Graphical Statistics (Matthew Simpson, Jarad Niemi, and Vivekananda Roy)
-  Bayesian Inference for a Covariance Matrix, Proceedings of the 26th Annual Kansas State University Conference on Applied Statistics in Agriculture. (Ignacio Alvarez-Castro, Matthew Simpson, and Jarad Niemi)
-  Application of Interweaving in DLMs to an Exchange and Specialization Experiment, Bayesian Statistics from Methods to Models and Applications: Research from BAYSM 2014. (Matthew Simpson)
Posters and Presentations
- Estimating Distributions for Populations Within Nested Geographies with Public-Use Data. Topic contributed presentation at JSM2017 in Baltimore, August 2, 2017
- Estimating Distributions for Populations Within Nested Geographies with Public-Use Data. ACS Data Users Conference in Alexandria, VA, May 12, 2017
- Introduction to Stan for Markov Chain Monte Carlo. Center for Survey Research and Methodology invited seminar at the U.S. Census Bureau in Suitland, MD, April 2017. Accompanying files.
- Estimating Distributions for Populations Within Nested Geographies with Public-Use Data. NSF-Census Research Network spring meeting in Suitland, MD, April 24, 2017
- Particle Swarm Optimization — Assisted Markov Chain Monte Carlo. Topic contributed presentation at JSM2016 in Chicago, August 3, 2016.
- Particle Swarm Optimization Assisted Markov Chain Monte Carlo. Poster at Spatial and Spatio-Temporal Design and Analysis for Official Statistics workshop at University of Missouri, May 20, 2016.
- Interweaving Markov Chain Monte Carlo Strategies for Efficient Estimation of Dynamic Linear Models. Presentation at BAYSM’14 at the WU Vienna University of Business and Economics, September 18, 2014.
- Ancillary-Sufficiency or not; Interweaving to Improve MCMC Estimation of the Local Level Model. Poster at the EFaB meeting of the Bayes 250 conference at Duke University, December 16, 2013.