My talk at MIT this Thursday

When I was a student at MIT, there was no statistics department. I took a statistics course from Stephan Morgenthaler and liked it. (I’d already taken probability and stochastic processes back at the University of Maryland; my instructor in the latter class was Prof. Grace Yang, who was super-nice. I couldn’t follow half of what was going on in that class, but she conveyed a sense of the mathematical excitement of the topic.) Statistics was still a bit mysterious to me. I thought I’d take another class but the math department didn’t offer much statistics and I didn’t want to take something boring. I asked my advisor who recommended I talk with Prof. Chernoff who recommended I take a course at Harvard. Which I did, but that’s another story.

Anyway, it seems that MIT is starting some sort of statistics program, and they invited me to the inaugural symposium! Which is cool.

My talk is this Thurs, 14 May, at 10am at room 46-3002. I’m pretty sure this is a new building.

Here are my title and abstract:

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

“Big Data” is more than a slogan; it is our modern world in which we learn by combining information from diverse sources of varying quality. But traditional statistical questions—how to generalize from sample to population, how to compare groups that differ, and whether a given data pattern can be explained by noise—continue to arise. Often a big-data study will be summarized by a little p-value. Recent developments in psychology and elsewhere make it clear that our usual statistical prescriptions, adapted as they were to a simpler world of agricultural experiments and random-sample surveys, fail badly and repeatedly in the modern world in which millions of research papers are published each year. Can Bayesian inference help us out of this mess? Maybe, but much research will be needed to get to that point.

(The title is long because I wasn’t sure which of 3 talks to give, so I thought I’d give them all.)

3 thoughts on “My talk at MIT this Thursday

  1. This talk sounds a bit hyper – “Every Parameter Would Like to Grow Up to Be a Distribution”

    Sorry, couldn’t help my self – many interesting speakers.

  2. > I’m pretty sure this is a new building.

    Yup. Relatively new at least. Brain and Cognitive Science used to be in what’s now E14 – on Amherst between Carleton and Ames.

    Sorry I’ll miss it but no flexibility in my schedule tomorrow AM.

Leave a Reply

Your email address will not be published. Required fields are marked *