BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multivariate power laws in a preferential attachment network model
 \; model calibration - Sidney Resnick (Cornell University)
DTSTART:20161214T160000Z
DTEND:20161214T164500Z
UID:TALK69498@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:We begin with a review of the multivariate regular variation o
 f in- and out-degree in a preferential attachment model. The problem can b
 e approached in a variety of ways: (i) Multivariate Tauberian theory\; (ii
 ) Direct approach via asymptotics to get a limit measure\; (iii) proving m
 ultivariate regular variation of the limiting mass&nbsp\;<b></b><b></b><b>
 </b><b></b>function of normalized in- and out-degree. We then turn to mode
 l calibration comparing various information sources and methods. If a full
  history of network growth is available\, full MLE implementation is possi
 ble and performs well on simulated data. If a single snapshot in time is a
 ll that is available\, then approximate MLE can be used. Comparison with M
 LE and use of asymptotic methods relying on heavy tail estimators is also 
 made and predictably there is a trade-off between robustness and accuracy.
  Methods generally perform well on simulated data but real data creates pr
 oblems with model error and we can illustrate this with wikipedia data obt
 ained from Konect.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR