BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Telerik Inc.//Sitefinity CMS 15.1//EN
BEGIN:VTIMEZONE
TZID:Eastern Standard Time
BEGIN:STANDARD
DTSTART:20251102T020000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYHOUR=2;BYMINUTE=0;BYMONTH=11
TZNAME:Eastern Standard Time
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250301T020000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYHOUR=2;BYMINUTE=0;BYMONTH=3
TZNAME:Eastern Daylight Time
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
DESCRIPTION:Yulong Wang\, associate professor of economics\, will present “
 Don’t Waste the Outliers in your Dataset – Theory and Application of Extre
 me Value Theory” as part of the CPR Methods Workshop Series.Extreme values
 \, such as extremely high medical expenditures and cost of catastrophic di
 sasters\, are common in practice. A routine method is to treat them as out
 liers and drop them from the data. However\, doing so could incur large bi
 as and substantially misleading statistic inference. In this talk\, we wil
 l discuss the effects of these extreme values on regressions and some alte
 rnatives that are robust to extreme values. Datasets about medical expendi
 ture\, city size and firm size will be illustrated.&nbsp\;
DTEND:20231012T203000Z
DTSTAMP:20260512T135039Z
DTSTART:20231012T193000Z
LOCATION:
SEQUENCE:0
SUMMARY:CPR Methods Workshop: Yulong Wang “Theory and Application of Extrem
 e Value Theory”
UID:RFCALITEM639141762397765450
X-ALT-DESC;FMTTYPE=text/html:<p>Yulong Wang\, associate professor of econom
 ics\, will present “Don’t Waste the Outliers in your Dataset – Theory and 
 Application of Extreme Value Theory” as part of the CPR Methods Workshop S
 eries.</p><p><strong></strong>Extreme values\, such as extremely high medi
 cal expenditures and cost of catastrophic disasters\, are common in practi
 ce. A routine method is to treat them as outliers and drop them from the d
 ata. However\, doing so could incur large bias and substantially misleadin
 g statistic inference. </p><p>In this talk\, we will discuss the effects o
 f these extreme values on regressions and some alternatives that are robus
 t to extreme values. Datasets about medical expenditure\, city size and fi
 rm size will be illustrated.&nbsp\;</p>
END:VEVENT
END:VCALENDAR