I am the John L. Loeb Associate Professor of the Natural Sciences at Harvard University, which actually means that I am a tenure-track faculty there. I develop methods for statistical inference, including Monte Carlo methods, techniques to compare models, to estimate latent variables and to deal with intractable normalizing constants. At Harvard I have taught courses on time series, statistical inference, and graduate seminars on particle filters, Bayesian nonparametrics and computational optimal transport.
In my recent articles, I look more specifically into:
the use of probability couplings to obtain unbiased estimators;
nonlinear dynamical systems and state space models;
statistical inference using ideas from optimal transport;
Bayesian inference in models made of modules, and model misspecification.
If you'd like to read one of my recent articles, I'd recommend Unbiased Markov chain Monte Carlo methods with couplings, with John O'Leary and Yves F. Atchadé, published with a discussion in JRSS B (2020).
I am working actively on lecture notes on "Couplings and Monte Carlo", available on this page.
Universität-Humboldt, Forschungsseminar Mathematische Statistik, Nov 4
Universität-Bonn, Oberseminar Stochastics, Nov 19
Rutgers University, Statistics seminar, December 2
Bayes Club, December 3
The up-to-date list of my research articles is on Google Scholar here.
My research is funded by the National Science Foundation, in grants DMS 1712872 and DMS 1844695 (CAREER).
You can find my CV here.
I am not taking interns over the summer. If you're interested in pursuing a PhD program, check the Department's website.
Email: pjacob at fas.harvard.edu.
Phone: (617) 496-9259
Office: Science Center 711, 1 Oxford Street, Cambridge, MA 02138