BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rothschild Lecture: From Small Data to Big Data and Back: Statisti
 cs and Data Science - Peter Bickel (University of California\, Berkeley)
DTSTART:20160817T150000Z
DTEND:20160817T160000Z
UID:TALK66954@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>  <span>Modern statistics began with R.A.Fisher&rsquo\;s
  seminal work in the early twentieth century\, with <a target="_blank" rel
 ="nofollow">important predecessors such as Karl Pearson\, contemporaries s
 uch as J.Neyman and successors </a>such as A. Wald. The focus was on small
  data sets or summaries of larger ones. Analyses were based on simple mode
 ls and given in terms of estimates\, testing and confidence bounds. </span
 >  &nbsp\;  With the advent of big data\, the size of datasets\, their com
 plexity and heterogeneity\, and the lack of theory to build mechanistic pr
 obability models brought new issues to the fore.&nbsp\;&nbsp\; I will disc
 uss some of these issues: </span>  1. Computation  2. Prediction  3. Spars
 ity/dimension reduction4. Stability/robustness   5. Reduction to small dat
 a
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR