I'm currently looking for a postdoctoral researcher and 2-3 doctoral students. Please contact me if you might be interested!
I'm an Assistant Professor in the Department of Statistics and Actuarial Science at the University of Waterloo, in Waterloo, Ontario, Canada. I'm originally from the UK, and moved to Canada in 2013 to work as a postdoc at McGill University. I received my PhD from the London School of Hygiene and Tropical Medicine in 2012, an MSc (in statistics) from University College London in 2008, and an MA (in mathematics) from Trinity College, Cambridge, in 2007.
My primary research interest is in causal inference, with a specific focus on dynamic treatment regimes (DTRs) and personalized medicine. Dynamic treatment regimes are sequences of decision rules that take subject-level data (such as age, health status, or prior treatment) as input and recommend actions (such as which drug to take) as output. Working with longitudinal datasets, my work focuses on deriving methodologies that help identify the sequence of treatment decisions that yields the best expected outcome.
More generally, I'm interested in identifying new ways to apply methods from different disciplines in new settings. This includes modifying methodology from one area of statistics so that it may be applied in a different area (such as applying measurement error techniques to dynamic treatment regime problems), or through applying statistical methods to novel problems in the 'real world' of data analysis.
Outside of research I enjoy 'recreational statistics' (i.e., fun problems that I haven't worked out how to get funding for yet). I'm also on the editorial board of Significance, the magazine of the American Statistical Association and the Royal Statistical Society. If you're unfamiliar with it do please take a look! We write accessible stories about how statistics is used in virtually every area of life. We're also always looking for contributors (both online and for the print magazine), so if you have an idea - no matter how thoroughly thought out - please feel free to get in touch!
Beyond statistics I'm a big fan of trivia (especially pub trivia nights), and have appeared on a number of game shows on British television. When not hiding from the Canadian winter, meanwhile, I like getting outdoors to go hiking, canoeing, and kayaking.
Some of my more recent and/or interesting papers are listed below (with a short explanation of each). If a paper isn't available online, please email me for a copy!
A couple of methodological papers:
- M P Wallace, E E M Moodie, D A Stephens, (2016). Model assessment in dynamic treatment regimen estimation via double robustness. Biometrics. 72(3): 855-864. [Link] Demonstrates how to use the double robustness property to assess nuisance models in G-estimation analyses.
- M P Wallace, E E M Moodie, (2015). Doubly-robust dynamic treatment regimen estimation via weighted least squares. Biometrics. 71(3): 636-644. [Link] Introduces a doubly-robust approach to DTR estimation via weighted ordinary least squares.
Some introductory/review articles:
- M P Wallace, E E M Moodie, (2014). Personalizing Medicine: A Review of Adaptive Treatment Strategies. Pharmacoepidemiology and Drug Safety. 23: 580-585. [Link] Introductory concepts and example analysis.
- M P Wallace, E E M Moodie, D A Stephens, (2016). SMART thinking: a review of recent developments in sequential multiple assignment randomized trials. Current Epidemiology Reports. [In Press] An overview of sequential multiple assignment randomized trials; a common study design for DTR analyses.
My R package for DTR estimation, DTRreg:
- DTRreg package on CRAN
- M P Wallace, E E M Moodie, D A Stephens, (2016). Dynamic treatment regimen estimation via regression-based techniques: Introducing R package DTRreg. Journal of Statistical Software. [Accepted pending minor revisions.] Introduces and demonstrates the DTRreg package.
...and now for something completely different:
- M P Wallace, C E Stewart, M J Moseley, D A Stephens, A R Fielder, (2013). Compliance with occlusion therapy for childhood amblyopia.Investigative Ophthalmology and Visual Science. 54: 6158-6166. [Link] Nothing to do with DTRs, but an interesting analysis of compliance with treatment among children with the eye condition amblyopia.