Notes on probabilistic modeling, and what we choose to model.

I’m a statistician working on uncertainty, dependence, and risk in complex systems — especially where data is thin and the extremes matter most.

Latest essays

Writing

Essays on probability, dependence, and modeling risk under real-world constraints.

Supervised Learning

The missing question in supervised learning

Supervised learning almost always predicts the mean. But the question that usually matters is the whole conditional distribution — what's possible, not just what's expected.

Feb 18, 2018

The Probaverse

An open-source ecosystem in R for treating probability distributions as tangible objects — the engine behind the way I think. It lets you explore the full space of possible outcomes in an analysis, rather than collapsing everything to a single number.

Explore the Probaverse

Selected Work

Applied projects where these ideas meet real systems.

Advances in Regional Flood Frequency Analysis (RFFA)

A collaboration with The University of British Columbia's Master of Data Science capstone course to advance methods in RFFA.

Jul 1, 2023

A Tale of Two Tributaries: Floodplain mapping in the Nicola Valley with Copulas

Modelling the design flood in the Nicola Valley when multiple mixed-regime watersheds are involved.

Jun 27, 2023

Error-correcting flow measurements from a level logger

Filling in the gaps between manually computed measurements of discharge by using kriging to error-adjust the automated measurements taken by a level logger.

Nov 30, 2022

See all

Get in touch

Working on a hard problem involving uncertainty or compound risk? I’m always glad to think it through. Reach out anytime.

Email me