BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:-//WordPress - MECv5.18.0//EN
X-ORIGINAL-URL:https://canssiontario.utoronto.ca/
BEGIN:VEVENT
UID:MEC-4ef42b32bccc9485b10b8183507e5d82@canssiontario.utoronto.ca
DTSTART:20220131T203000Z
DTEND:20220131T213000Z
DTSTAMP:20210310T193400Z
CREATED:20210310
LAST-MODIFIED:20220113
SUMMARY:Data Science ARES: Ben Bolker
DESCRIPTION:\nJoin us at the Data Science Applied Research and Education Seminar (ARES) with:\n\n\n\nDr. Benjamin BolkerDirector, School of Computational Science and EngineeringProfessor, Department of Mathematics & Statistics and Department of BiologyActing Associate Chair (Graduate), MathematicsMcMaster University\n\n\n\nFree Event | Registration Required\n\n\n\nTalk Title: No free lunch in inference\n\n\n\nAbstract: \n\n\n\nStatistical methods can target exploration, prediction, or inference.While big-data applications have emphasized prediction, inferenceremains important; in particular, inference is closely related toassessing the uncertainty of coefficients and predictions. Data-drivenmethods for model selection and tuning minimize prediction error bytrading bias for variance, but they are rarely (never?) able to narrowconfidence intervals or increase certainty. If used naively, popularmethods of data-driven model selection and tuning lead tooverconfidence. Post-selection inference, a non-naive method ofaccounting for the effects of data-driven model tuning, rely on strongassumptions. Researchers should should recognize how hard it is toquantify uncertainty reliably when they use data-driven model tuning,and in many cases should abstain from tuning altogether.\n\n\n\nSpeaker Profile: \n\n\n\nDr. Benjamin Bolker completed an undergraduate degree in mathematics and physics at Yale University and a Ph.D. in Zoology at CambridgeUniversity, working on the dynamics of measles epidemics. He did apostdoc at Princeton University in ecology and evolutionary biology onspatial dynamics of plant and host-parasite communities, beginning afaculty position at the Department of Zoology (later Biology) at theUniversity of Florida in 1999. He moved to McMaster University in 2010,where he has a joint appointment in Mathematics & Statistics and Biologyand directs the School of Computational Science and Engineering. Hisresearch ranges broadly across ecology, evolution, and epidemiology,applying mathematical, statistical, and computational tools. He isespecially interested in problems that involve parasites and disease,spatial population dynamics, estimation and inference of modelparameters from observational data, or all three. In addition to manyresearch papers, he is the author of two books (Ecological Models and Data in R and A Very Short Introduction to Infectious Disease, withMarta Wayne) and the author or maintainer of several widely used R packages.\n
URL:https://canssiontario.utoronto.ca/event/ares_ben_bolker/
ORGANIZER;CN=Esther Berzunza:MAILTO:esther.berzunza@utoronto.ca
CATEGORIES:Applied Research and Education Seminarâ€‹
LOCATION:Zoom (Online)
ATTACH;FMTTYPE=image/png:https://canssiontario.utoronto.ca/wp-content/uploads/2021/03/ben_bolker_3x2.png
END:VEVENT
END:VCALENDAR