I am a 5th year PhD student in the Statistics Department at the University of Michigan. I research likelihood based methods for partially observed Markov process (POMP) models. I primarily work on theory, methodology, and software related to these models. My application area has been modeling infectious disease outbreaks, but I am also interested in a variety of other topics and application areas.
My thesis advisor is Edward Ionides.
Education
University of Michigan | Ann Arbor, MI PhD in Statistics | August 2020 - Current
Utah State University | Logan, UT B.S. in Mathematics and Statistics | August 2016 - May 2020
Selected Papers
Wheeler, J., et al. 2024. “Informing policy via dynamic models: Cholera in Haiti”. PLOS Computational Biology.
Wagstaff, J., Bean, B., Wheeler, J., Maguire, M., Sun, Y. 2024. “Adaptive Mapping of Design Ground Snow Loads in the Conterminous United States”. Journal of Structural Engineering.
Wheeler, J., Ionides, E. L. 2023. “Likelihood Based Inference of ARMA Models”. ArXiv preprint. arXiv.2310.01198.
Ionides, E. L., Ning, N., Wheeler, J. 2024. “An Iterated Block Particle Filter for Inference on Coupled Dynamic Systems with Shared and Unit-Specific Parameters”. Statistica Sinica.
Wheeler, J., Bean, B., Maguire, M. 2022. “Creating a universal depth-to-load conversion technique for the conterminous United States using random forests”. Journal of Cold Regions Engineering.
Software
- arima2: This library aids maximum likelihood estimation of parameters of ARIMA time series models. The package is currently on CRAN, and there is an associated pre-print paper on ArXiv.
- panelPomp: This R package on CRAN is used for inference of Panel POMP models. I am the current maintainer of this package, and creator and admin for the corresponding GitHub organization.
This website was created using Quarto. To learn more about Quarto websites, see slides I created for a Statistics Student Seminar at the University of Michigan, 2024.
