event

21st-Century Statistics at MIT: Inaugural Symposium
May 14, 2015 - May 15, 2015 | Singleton Auditorium (Building 46-3002)

The vast amount of data that is now available from physical, engineered, and social systems is revolutionizing the field of statistics. In many cases, the availability of more data creates more opportunities to understand, model, and control complex systems. At the same time, new challenges emerge in the need to analyze and interpret large, heterogeneous data sets in ways that can improve systems and policies.

MIT’s new Center for Statistics, part of the MIT Institute for Data, Systems, and Society, hosted the two-day symposium, “21st-Century Statistics,” on May 15-16. The event offered technical presentations by thought leaders in mathematical statistics, machine learning, econometrics, and biostatistics—providing the MIT community with a forum for discussing some of the challenges and opportunities that are redefining the field of statistics. (Agenda, speaker info, and other details at statsconf.mit.edu.)

Click on the talks below to view videos:


Welcome from the event organizers; a look at the new MIT Institute for Data, Systems, and Society – Munther Dahleh, Massachusetts Institute of Technology


“On the Detection of Non-Independence” – Jun Liu, Harvard University


“Little Data: How Traditional Statistical Ideas Remain Relevant in a Big-Data World; or, The Statistical Crisis in Science; or, Open Problems in Bayesian Data Analysis” – Andrew Gelman, Columbia University


“Statistical Thinking in Neuroscience” – Rob Kass, Carnegie Mellon University


“Statistical Analysis Inside and Outside Economic Models” – Lars Peter Hansen, University of Chicago


“Micro-Randomized Trials & mHealth” – Susan Murphy, University of Michigan


“Statistics and Computation in Graph Estimation Problems” – Andrea Montanari, Stanford University


“Bayes, De Fenetti, and the Challenges of Statistical Inference for Documents” –
Michael Jordan, Berkeley College


“What Can We Learn From Asymptotic Properties of the Posterior
Distributions in Large or Infinite Dimensional Models?” – Judith Rousseau, University of Paris


© MIT Institute for Data, Systems, and Society | 77 Massachusetts Avenue | Cambridge, MA 02139-4307 | 617-253-1764 | Design by Opus