BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:CAM Colloquium: Sumanta Basu (Statistics and Data Science & Co
 mputational Biology\, Cornell) - Measuring systemic risk with graphical mo
 dels of time series data
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260719T124211Z
UID:tag:localist.com\,2008:EventInstance_31309948432175
DTSTART:20191025T193000Z
DTEND:20191025T203000Z
DESCRIPTION:Abstract:\nAfter the 2007-09 financial crisis\, there has been 
 a growing interest in measuring systemic risk\, broadly defined as the ris
 k of widespread failure of the entire financial system. In a highly interl
 inked financial market\, a large body of recent works have proposed to use
  network connectivity amongst financial institutions to assess their syste
 mic importance. In this work\, we will present some graphical modeling tec
 hniques for learning interactions among the  components of a large dynamic
  system from multivariate time series data\, where the core idea is to lea
 rn from lead-lag relationships (commonly known as Granger causality) betwe
 en time series in addition to their co-movements. In the context of modeli
 ng networks of interactions amongst financial institutions and measuring s
 ystemic risk\, we will demonstrate how linear and quantile-based Granger c
 ausality analyses using vector autoregressive (VAR) models can provide ins
 ight. We will present some non-asymptotic statistical theory for our propo
 sed algorithms\, estimate these graphical models using stock returns of la
 rge financial institutions in U.S. and India\, and demonstrate their usefu
 lness in detecting systemically risky periods and institutions. This is jo
 int work with Kara Karpman and Diganta Mukherjee.\n\nBio:\nI am broadly in
 terested in developing statistical machine learning methods for structure 
 learning and prediction of complex\, high-dimensional systems arising in b
 iological and social sciences. I am currently working in two areas: (a) ne
 twork modeling of high-dimensional time series\; and (b) detecting high-or
 der interactions in complex biological systems using randomized tree ensem
 bles. I also work closely with scientists and economists on a wide range o
 f problems including prostate cancer progression\, large scale metabolomic
 s\, and systemic risk monitoring in financial markets. Before joining Corn
 ell\, I was a postdoctoral scholar (2014-2016) in the Department of Statis
 tics\, UC Berkeley and the Biosciences Division\, Lawrence Berkeley Nation
 al Laboratory . I received my PhD (2014) from the Department of Statistics
 \, University of Michigan \, and my bachelors (2006) and masters (2008) in
  Statistics from Indian Statistical Institute\, Kolkata.
GEO:42.443451;-76.481506
LOCATION:Frank H. T. Rhodes Hall\, 655
SUMMARY:CAM Colloquium: Sumanta Basu (Statistics and Data Science & Computa
 tional Biology\, Cornell) - Measuring systemic risk with graphical models 
 of time series data
URL;VALUE=URI:https://events.cornell.edu/event/cam_colloquium_sumanta_basu_
 statistics_and_data_science_computational_biology_cornell_-_measuring_syst
 emic_risk_with_graphical_models_of_time_series_data
CATEGORIES:Colloquium
END:VEVENT
END:VCALENDAR