Rachel Carrington

Rachel Carrington

Research Associate, University of Bath

About me

I am currently a Research Associate in Statistics at the University of Bath, where I am part of the Centre of Excellence in Water-Based Early Warning Systems for Health Protection (CWBE).

Previously, I worked with Paul Fearnhead at Lancaster University on post-selection inference for changepoint models. This involved adapting existing methods to obtain greater power in detecting changes, and developing methods for post-selection inference in the piecewise constant variable model.

I obtained my MMath and PhD from the University of Nottingham, the latter under the supervision of Karthik Bharath and Simon Preston. My PhD thesis focused on word embeddings used to model language. I investigated several aspects of this topic, including addressing the non-identifiability of sets of word embeddings, and considering methods for modelling language change across time. I followed this with a postdoc at the University of Nottingham, where I worked on combining high-dimensional, non-traditional forms of data to map urban deprivation in Tanzania.

Contact: rjc204 (at) bath.ac.uk

Research Interests

  • Wastewater epidemiology
  • Changepoint detection
  • High-dimensional data
  • Post-selection inference

For more details about my research and publications, click on the “Research” tab above.

Publications

Urban mapping in Dar es Salaam using AJIVE. Rachel Carrington, Ian Dryden, Madeleine Ellis, James Goulding, Simon Preston, and David Sirl. To appear in Journal of the Royal Statistical Society: Series C – Applied Statistics. link to arXiv version

Improving Power by Conditioning on Less in Post-selection Inference for Changepoints. Rachel Carrington and Paul Fearnhead (2025). Statistics and Computing 35(8). link

Invariance and identifiability issues for word embeddings. Rachel Carrington, Karthik Bharath and Simon Preston (2019). Advances in Neural Information Processing Systems (NEURIPS) 32. link

Preprints

Post-selection inference for quantifying uncertainty in changes in variance. Rachel Carrington and Paul Fearnhead. Under review. arXiv preprint. link