My dissertation research focused on heterogeneity estimation in challenging generative scenarios: imagine high-order effects that you either don’t know or can’t enumerate, or synthesizing learnings from multiple, incomplete data sets that suffer from missingness. A selection of pre-prints and publications are listed below.



Partitioning Marginal Epistasis Distinguishes Nonlinear Effects from Polygenicity and Other Biases in GWA Summary Statistics
Darnell, G., Smith, S.P., Udwin, D., Ramachandran, S., Crawford, L. (2022)
bioRxiv.

A Group Variable Importance Framework for Bayesian Neural Networks
Ish-Horowicz*, J., Udwin*, D., Kolbeinsson, A., Scharfstein, K., Flaxman, S., Crawford\(\dagger\), L., Filippi\(\dagger\), S. (2019)
arXiv.
*Authors contributed equally;
\(\dagger\)Authors contributed equally

R Markdown
Baumer, B., Udwin, D. (2015)
WIREs: Computational Statistics.

What Percent of the Continental US is Within One Mile of a Road?
Stoudt, S., Cao, Y., Udwin, D., Horton, N.J. (2014)
STatistics Education Web.