Ordinary Meeting Organised by the Environmental Statistics Section
Geostatistical inference under preferential sampling
P. J. Diggle (Lancaster University and Johns Hopkins University
School of Medicine, Baltimore), R. Menezes (University of Minho) and T.
Su (Lancaster University)
Wednesday 23 September 2009, 5-7pm to be followed by a free wine
reception.
The meeting will be preceded by a Pre-Ordinary meeting at 3pm. See
http://www.maths.ed.ac.uk/~nbochkin/RSSpreordinary.htm
for details.
RSS Headquarters, 12 Errol Street, London
Summary
Geostatistics involves the fitting of spatially continuous models to
spatially discrete data. Preferential sampling arises when the process
that determines the data-locations and the process being modelled are
stochastically dependent. The paper proposes an idealised model for
preferentially sampled geostatistical data, uses the model to show how
ignoring preferential sampling can lead to misleading inferences, and
describes an application to a set of bio-monitoring data from Galicia,
northern Spain.
The paper can be downloaded in advance from:
http://www.rss.org.uk/PDF/DiggleSept09.pdf
Fellows unable to attend the meeting itself are welcome to submit
written comments on the paper by 7 October; if submitted before 23
September they may be read during the meeting. All such contributions
must be no longer than 400 words and should be submitted to Charlotte
Stovell ([log in to unmask]).
No registration is required for this event.
There is no charge for this event.
************************************************************
Richard Wilkinson - Lecturer in Statistics
School of Mathematical Sciences
University of Nottingham
[log in to unmask]
************************************************************
|