Peer-reviewed papers
Note:The links for "Time capsuled code" provide code for reproducing the numerical results of each paper.
† Graduate student working under my supervision
∗ Undergraduate student working under my supervision
Foundations of statistics
- N Dey†, R Martin, and J P Williams (202x). Anytime-valid generalized universal inference on risk minimizers. In review.
Link: Manuscript. - J P Williams (202x). Model-free generalized fiducial inference. R&R at Journal of Machine learning Research.
Link: Manuscript. Link: Time capsuled code (Julia code). - N Dey† and J P Williams (202x). Valid inference for machine learning model parameters. In review.
Link: Manuscript. - A Murph†, J Hannig, and J P Williams (202x). Generalized fiducial inference on differentiable manifolds. In review.
Link: Manuscript. - I Carmichael and J P Williams (2018). An exposition of the false confidence theorem. Stat 7 (1) pp.e201.
Link: Manuscript. Link: Time capsuled code (R code).
Machine learning
- A Hjort†, J P Williams, and J Pensar (202x). Clustered conformal prediction for the housing market. In review.
- A Hjort†, G H Hermansen, J Pensar, and J P Williams (202x). Uncertainty quantification in automated valuation models with locally weighted conformal prediction. In review.
Link: Manuscript. - A Murph†, C B Storlie, P M Wilson, , J P Williams, and J Hannig (202x). Bayes Watch: Bayesian Change-point Detection for Process Monitoring with Fault Detection. In review.
Link: Manuscript. - N Dey†, M Singer†, J P Williams, and S Sengupta (202x). Word Embeddings as Statistical Estimators. To appear in Sankhya B.
Link: Manuscript. - N Dey†, J Ding†, J Ferrell∗, C Kapper∗, M Lovig∗, E Planchon∗, and J P Williams (2023). Conformal prediction for text infilling and part-of-speech prediction. New England Journal of Statistics in Data Science 1 pp.69--83.
Link: Manuscript. Link: Time capsuled code (Julia code and Python code).
Multistate Markov and hidden Markov models
- J P Williams, G H Hermansen, H Strand, G Clayton, and H M Nygård (202x). Bayesian hidden Markov models for latent variable labeling assignments in conflict research: application to the role ceasefires play in conflict dynamics. To appear in Annals of Applied Statistics.
Link: Manuscript. Link: Time capsuled code (R code). - E B Kendall†, J P Williams, G H Hermansen, F Bois, and V H Thanh (202x). Beyond time-homogeneity for continuous-time multistate Markov models. R&R at Journal of Computational and Graphical Statistics.
Link: Manuscript.
Description: Extending the ideas of the paper below, this research establishes how to fit time-inhomogeneous, continuous-time HMM, not restricted to the assumption that transition rates are piecewise constant functions of time. The approach relies on numerical solutions to the Kolmogorov forward equations, and doesn't suffer from the biases that we demonstrate for the matrix-exponential solution applied (as is commonly done) when the transition rates are not piecewise constant.
- J P Williams, C B Storlie, T M Therneau, C R Jack Jr, and J Hannig (2020). A Bayesian approach to multi-state hidden Markov models: application to dementia progression. Journal of the American Statistical Association 115 (529) pp.16-31.
Link: Manuscript. Link: Supplementary material. Link: Time capsuled code (R code).
Description: Primarily, this research determines how to fit a time-inhomogeneous, continuous-time HMM, based on the assumption that the transition rates are piecewise constant functions of time. The approach is exact, and relies on the matrix-exponential solution to the Kolmogorov forward equations.
Model selection
- S Koner† and J P Williams (2023). The EAS approach to variable selection for multivariate response data in high-dimensional settings. Electronic Journal of Statistics 17 (2) pp.1947--1995.
Link: Manuscript. Link: Time capsuled code (R code).
Description: Third paper on the EAS approach to model selection. - J P Williams, Y Xie, and J Hannig (2023). The EAS approach for graphical selection consistency in vector autoregression models. Canadian Journal of Statistics 51 (2) pp.674--703.
Link: Manuscript. Link: Supplementary material. Link: Time capsuled code (mostly Python code, some R code).
Description: Second paper on the EAS approach to model selection. - J P Williams and J Hannig (2019). Nonpenalized variable selection in high-dimensional linear model settings via generalized fiducial inference. Annals of Statistics 47 (3) pp.1723-1753.
Link: Manuscript. Link: Supplementary material. Link: Time capsuled code (mostly Python code, some R code).
Description: First paper on the epsilon-admissible subsets (EAS) approach to model selection. - J P Williams, D M Ommen, and J Hannig (2023). Generalized fiducial factor: an alternative to the Bayes factor for forensic identification of source problems. Annals of Applied Statistics 17 (1) pp.378–402.
Link: Manuscript. Link: Time capsuled code (R code). - J P Williams and Y Lu (2015). Covariance Selection in the Linear Mixed Effect Model. Journal of Machine Learning Research: Workshop and Conference Proceedings 44 pp.277-291. (NIPS conference session).
Link: Manuscript. Link: Time capsuled code (R code).
Transfer learning
- M A Abba†, J P Williams, and B J Reich (202x). A Bayesian shrinkage estimator for transfer learning. In review.
Link: Manuscript. - J Hickey†, J P Williams, and E C Hector (202x). Transfer Learning with Uncertainty Quantification: Random Effect Calibration of Source to Target (RECaST). R&R at Journal of Machine learning Research.
Link: Manuscript. - M A Abba†, J P Williams, and B J Reich (2023). A penalized complexity prior for deep Bayesian transfer learning with application to materials informatics. Annals of Applied Statistics 17 (4) pp.3241–3256.
Link: Manuscript.
Discussions/chapters appearing in journals/books with statistical focus
- N Dey†, R Martin, and J P Williams (202x). Neil Dey, Ryan Martin, and Jonathan P Williams' contribution to the Discussion of "Safe Testing" by Grünwald, de Heide, and Koolen. To appear in Journal of the Royal Statistical Society: Series B.
- A Murph†, J Hannig, and J P Williams (2024). Introduction to generalized fiducial inference. Handbook of Bayesian, Fiducial, and Frequentist Inference (1st ed.). Chapman and Hall/CRC.
Link: Manuscript. Link: Time capsuled code (R code). - J P Williams (2021). Discussion of
A Gibbs sampler for a class of random convex polytopes
. Journal of the American Statistical Association 116 (535) pp.1198-1200.
Link: Manuscript.
Manuscripts appearing in journals with non-statistical focus
- V V Volpe, E B Kendall†, A N Collins, M G Graham, J P Williams, and S J Holochwost (202x). Prior exposure to racial discrimination and patterns of acute parasympathetic nervous system responses among black adults. In review.
- N Giertych†, A Shaban, P Haravu∗, and J P Williams (202x). A statistical primer on classical
period-finding techniques in astronomy. In review.
Link: Manuscript. Link: Time capsuled code (Python and R code). - I Carmichael, T Keefe, N Giertych†, and J P Williams (202x). yaglm: a Python package for fitting and tuning generalized linear models that supports structured, adaptive and non-convex penalties. In progress.
Link: Manuscript. - S Nghiem, J P Williams, C Afoakwah, Q Huynh, S K Ng, and J Byrnes (2021). Can administrative health data improve the gold standard? Evidence from a model of the progression of myocardial infarction. International Journal of Environmental Research and Public Health 18 (14) pp.7385.
Link: Manuscript. - E Sechi, E Shosha, J P Williams, S J Pittock, B G Weinshenker, B M Keegan, N L Zalewski, A S Lopez-Chiriboga, J Jitprapaikulsan, and E P Flanagan (2019). Aquaporin-4 and MOG autoantibody discovery in idiopathic transverse myelitis epidemiology. Neurology 93 (4) pp.e414-e420.
Link: Manuscript.