BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20151120T123200Z
DTSTART:20151215T160000Z
DTEND:20151215T173000Z
SUMMARY:CMIST afternoon seminar: Domain Prediction of Complex Indicators
– Model-based Methods and Robustness
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}plf-ih7nc9i
w-nnxrxp
DESCRIPTION:Small Area (Domain) prediction of complex indicators for exam
ple\, deprivation and inequality indicators typically relies on micro-si
mulation/model-based methods that use regression models with domain-spec
ific random effects. When standard (Gaussian) assumptions for the model
error terms are met\, Empirical Best Prediction (EBP) for domains is pos
sible and should be preferred. In this talk we will present current rese
arch on alternative methodologies when the model assumptions are possibl
y violated. To start with\, we will discuss the use of ‘optimal’ transfo
rmations for trying to ensure the validity of the assumptions required f
or EBP. We will then outline alternative\, possibly robust model-based m
ethodologies. These methods are based on the use of a random effects mod
el for the quantiles of the target distribution. By using such a model o
ne can estimate the quantile function of the target distribution which i
n turn can be used for micro-simulating samples to be used in domain pre
diction. The link between maximum likelihood estimation and the use of t
he Asymmetric Laplace Distribution as a working assumption will be descr
ibed. The proposed method can be used both with continuous and discrete
(count) outcomes. The talk will briefly outline future work on the use o
f this method with discrete outcomes in particular on how to impose smoo
thing for estimating the quantiles of discrete outcomes. Estimation of t
he Mean Squared Error of the domain parameters will be discussed and ope
n areas for research will be described. Some results by using simulation
studies and real data will be presented.
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:2.07\, Humanities Bridgeford Street\, Manchester
END:VEVENT
END:VCALENDAR