BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20181104T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:683908e46353d
DTSTART;TZID=America/Toronto:20191011T103000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20191011T103000
URL:/statistics-and-actuarial-science/events/distinguis
 hed-lecture-paul-glasserman-columbia-university
SUMMARY:Distinguished lecture by Paul Glasserman\, Columbia University
CLASS:PUBLIC
DESCRIPTION:Summary \n\nPRECISION FACTOR INVESTING: AVOIDING FACTOR TRAPS B
 Y PREDICTING\nHETEROGENEOUS EFFECTS OF FIRM CHARACTERISTICS\n\n-----------
 --------------\n\nWe apply ideas from causal inference and machine learnin
 g to estimate\nthe sensitivity of future stock returns to observable chara
 cteristics\nlike size\, value\, and momentum. By analogy with the informal
  notion of\na \"value trap\,\" we distinguish \"characteristic traps\" (st
 ocks with\nweak sensitivity) from \"characteristic responders\" (those wit
 h strong\nsensitivity). We classify stocks by interpreting these distincti
 ons as\nheterogeneous treatment effects (HTE)\, with characteristics\ninte
 rpreted as treatments and future returns interpreted as responses.\nThe cl
 assification exploits a large set of stock features and recent\nwork apply
 ing machine learning to HTE. Long-short strategies based on\nsorting stock
 s on characteristics perform significantly better when\napplied to charact
 eristic responders than traps. A strategy based on\nthe difference between
  these long-short returns profits from the\npredictability of HTE rather t
 han from factors associated with the\ncharacteristics themselves. This is 
 joint work with Pu He.\n 
DTSTAMP:20250530T012452Z
END:VEVENT
END:VCALENDAR