Category Theory
Zulip Server
Archive

You're reading the public-facing archive of the Category Theory Zulip server.
To join the server you need an invite. Anybody can get an invite by contacting Matteo Capucci at name dot surname at gmail dot com.
For all things related to this archive refer to the same person.

Stream: event: Categorical Probability and Statistics 2020 workshop

Topic: Jun 5: Peter McCullagh's talk

Paolo Perrone (Jun 04 2020 at 15:15):

Hey all,
This is the discussion thread of Peter McCullagh's talk, "Categorical notions in statistics".
The talk, besides being on Zoom, is livestreamed here: https://youtu.be/fVEucjkYSmw

Paolo Perrone (Jun 04 2020 at 19:24):

Date and time: Friday, 5 Jun, 15h UTC.

Paolo Perrone (Jun 05 2020 at 14:30):

Hi. We will start in 30 minutes.

Paolo Perrone (Jun 05 2020 at 15:08):

Unfortunately the speaker is not connected yet. Apologies for the inconvenience!

Paolo Perrone (Jun 05 2020 at 15:16):

We are having a break for now, with breakout rooms. We will start again in about 40 minutes. Sorry about that!

Simon Burton (Jun 05 2020 at 15:17):

The paper mentioned by Tom: "WHAT IS A STATISTICAL MODEL?" by PETER MCCULLAGH http://www.stat.uchicago.edu/~pmcc/pubs/AOS023.pdf

Manuel Bärenz (Jun 05 2020 at 15:29):

Can you say that on Youtube as well?

Manuel Bärenz (Jun 05 2020 at 15:29):

(That we are having a break)

Paolo Perrone (Jun 05 2020 at 15:30):

Yep, thank you

Paolo Perrone (Jun 05 2020 at 15:37):

Ok, the talk is postponed. It is going to happen in 90 minutes, after Malte's talk!

Paolo Perrone (Jun 05 2020 at 16:55):

Starting in 5 minutes.

Tom Leinster (Jun 05 2020 at 17:32):

Will Peter's slides be available?

eric brunner (Jun 05 2020 at 17:32):

thanks!

eric brunner (Jun 05 2020 at 17:35):

i hope so. did you catch peter's remarks re: saunders? i missed those minutes.

Paolo Perrone (Jun 05 2020 at 17:36):

t.eric.brunner said:

i hope so. did you catch peter's remarks re: saunders? i missed those minutes.

While we wait for the slides (and possibly the recording), you can connect to the Youtube stream and watch from an earlier moment, if you want.

eric brunner (Jun 05 2020 at 17:42):

thanks. i will, hours from now.

Manuel Bärenz (Jun 05 2020 at 17:48):

I didn't really want to ask in the talk because it's probably more of an open discussion: Is there a relation between these statistics and higher inductive types? In particular, for every natural number n there is a higher inductive type corresponding to the groupoid given by the symmetric group of order n.

eric brunner (Jun 05 2020 at 18:02):

i'd not realized -- pages 1267-1310 are discussion of peter's "what is" paper, which begins at 1225.

Tobias Fritz (Jun 05 2020 at 20:29):

The slides are now here.

Alex Simpson (Jun 05 2020 at 21:04):

I enjoyed your talk, Peter. Isn't being a natural statistic for an S.R.S. a Martingale condition? Consider the Markov process whose states are all finite samples of size $\geq d$, with transitions from size $n$ to size $n-1$ defined by forgetting an element with uniform probability. Then the property of being a natural statistic says that $(T_n)$ is a Martingale with respect this Markov chain. I realise this is an obvious restatement (if true at all - it's late at night here and I may be missing something). But assuming it is correct, I find it helpful to see the definition in familiar terms.

Sam Staton (Jun 05 2020 at 21:12):

Tobias Fritz said:

The slides are now here.

Thanks -- I guess you meant https://perimeterinstitute.ca/personal/tfritz/2019/cps_workshop/slides/mccullagh.pdf .

Tobias Fritz (Jun 05 2020 at 21:34):

D'oh, yes thanks, corrected.

Paolo Perrone (Jun 05 2020 at 23:56):

Here's the recording!
https://youtu.be/FTYLuzRxHTs