My main research areas are firmly within the realm of Bayesian statistics and Markov chain Monte Carlo methods. I’m generally interested in models for dependent data, especially time series, spatial, and spatial temporal models, as well as MCMC and other computing methods for these models. I also have an interest in constructing priors for covariance matrices, partially motivated by the sorts of models I’m interested in. Finally, I’m also interested in applications of Bayesian statistics to causal inference, particularly in economics.
- [To Appear] 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
- 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.
- 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.