Jeffrey S. Rosenthal

Also known as Jeff Rosenthal

Jeffrey Rosenthal is a statistician at the University of Toronto who studies Markov chain Monte Carlo (MCMC) algorithms, the sampling methods that bring Bayesian inference to a multitude of settings. Born in Scarborough, Ontario, he earned a BSc at Toronto in 1988 and a Harvard PhD under Persi Diaconis in 1992, then joined Toronto's statistics department in 1993. With Gareth Roberts, he proved an optimal acceptance rate of 0.574 for Langevin-based Metropolis–Hastings proposals in 1998 and surveyed Markov chains on general state spaces in 2004. His 2005 book Struck by Lightning became a Canadian bestseller, and he won the 2006 CRM-SSC Prize and the 2007 COPSS Presidents' Award. A Vector Institute Faculty Affiliate since 2019, he is a professor in Toronto's Department of Statistical Sciences.

Now at: University of Toronto and Vector Institute
Born: 1967
Badges: Boomerang
Loyal

Years link to their sources.

Timeline

1988–1992

Toronto and Harvard degrees

Born in Scarborough, Ontario in 1967, he earns a BSc in mathematics, physics and computer science from the University of Toronto in 1988. He completes a PhD in mathematics at Harvard in 1992 under Persi Diaconis, with a thesis titled Rates of Convergence for Gibbs Sampler and Other Markov Chains.
1992–1997

Joining Toronto's statistics department

He is an assistant professor in the Department of Mathematics at the University of Minnesota from 1992 to 1993. In 1993, he starts as an assistant professor in the University of Toronto's Department of Statistics, and he becomes an associate professor there in 1997.
1998–2004

Optimal scaling and a survey

In 1998, he and Gareth Roberts publish "Optimal scaling of discrete approximations to Langevin diffusions" in the Journal of the Royal Statistical Society, Series B, proving a diffusion limit and an asymptotically optimal acceptance rate of 0.574 for Metropolis–Hastings algorithms built on Langevin diffusions. In 2004, they publish the survey "General state space Markov chains and MCMC algorithms" in Probability Surveys, which presents conditions for geometric and uniform ergodicity and bounds on convergence rates.
2005–2006

Struck by Lightning and a lottery scandal

In 2005, he writes the general-audience book Struck by Lightning: The Curious World of Probabilities, a bestseller in Canada. In 2006, his statistical analysis for CBC's The Fifth Estate helps expose the Ontario lottery retailer fraud scandal.
2006–2013

Prizes in statistics

He wins the 2006 CRM-SSC Prize and the 2007 COPSS Presidents' Award. He is elected a Fellow of the Royal Society of Canada in 2012 and receives the SSC Gold Medal in 2013.
April 2017

The Magic of Monte Carlo

In April 2017, he gives the Alan B. Pritsker Scholars Distinguished Lecture at Purdue's School of Industrial Engineering, titled "The Magic of Monte Carlo". The talk uses Java applets to illustrate how MCMC algorithms, which have allowed Bayesian inference in a multitude of settings, work.
2019

Fujitsu lab and Vector Institute

In 2019, he is appointed a principal investigator of the Fujitsu Co-Creation Research Laboratory and a Faculty Affiliate of the Vector Institute for Artificial Intelligence.
As of October 2026

Professor of statistical sciences

He is a professor in the Department of Statistical Sciences at the University of Toronto, cross-appointed to the Department of Mathematics, and remains a Faculty Affiliate of the Vector Institute. His specialty is Markov chain Monte Carlo algorithms.