BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.bham.ac.uk//v3//EN
BEGIN:VEVENT
CATEGORIES:Astrophysics Seminars
SUMMARY:The Unified Statistical Analysis of Populations of
Sources: Advantages of “Shrinkage Estimates” in A
stronomy - David van Dyk (Imperial)
DTSTART:20150429T133000Z
DTEND:20150429T143000Z
UID:TALK1520AT
URL:/talk/index/1520
DESCRIPTION: \nAstronomical studies often involve samples or p
opulations of sources. The parameters describing t
he sources can either be fit to each source in a s
eparate analysis\, or all be fit in a single unifi
ed analysis. The latter strategy allows us to inco
rporate the population distribution into a coheren
t statistical model and exhibits distinct statisti
cal advantages. In particular\, objects with small
er error bars and well-constrained parameters allo
w us to estimate the population distribution\, whi
ch in turn can be used to better estimate the weak
ly-constrained parameters associated with objects
with larger error bars. The fitted values of such
weakly-constrained parameters will “shrink towards
’’ the population mean\, and are thus called “shri
nkage estimates’’. This talk describes both frequ
entist and Bayesian advantages of shrinkage estima
tes and illustrates how they can be used in astron
omy. In the first of two examples we estimate the
absolute magnitudes of a SDSS sample of 288 Type I
a Supernovae using shrinkage estimates and illustr
ate how they differ from naive estimates. In the s
econd example\, we use photometric magnitudes of a
sample of galactic halo white dwarfs to simultane
ously obtain shrinkage estimates of the stellar ag
es and an estimate the age of the halo.
LOCATION:Physics West 117
CONTACT:Ilya Mandel
END:VEVENT
END:VCALENDAR