On behalf of Michael Greenacre, I forward the following course
announcement to the list:
=============================================================================
4th APPLIED STATISTICS WEEK
Short Courses
Statistics in the Environmental Sciences
Barcelona, 26 June to 2 July 1998
Presenters: Vic Barnett, Cajo ter Braak and Peter Diggle
Aims
The fourth APPLIED STATISTICS WEEK is being organized by Pompeu Fabra
University (UPF) from 26th June to 2nd July 1998, in Barcelona.
The APPLIED STATISTICS WEEK aims to provide a set of intensive courses
on a particular statistical theme of an applied nature. The courses
are presented by acknowledged leading researchers of international stature,
who are also known to have excellent teaching and communication skills.
Past themes of the APPLIED STATISTICS WEEK have been "Statistics in the
Health Sciences" (1995), "Statistics in Classification and Pattern
Recognition" (1996) and "Design and Analysis of Survey Data" (1997).
The theme in 1998 has been chosen as "Statistics in the Environmental
Sciences" and again three courses will be offered covering the key aspects
of this subject.
The principal aim of these courses is to improve the ability of researchers
to collect and analyze data in studies of the environment and of ecosystems.
The courses deal firstly with issues of sampling in environmental studies
and efficient sampling designs, secondly with a class of multivariate
techniques specially developed to analyze environmental and ecological data.
Since environmental data are often collected over a geographical area, the
third course deals with methods for modelling spatial data. After attending
these courses, participants will have a thorough experience, both from a
theoretical and a practical viewpoint, of these three issues in environ-
mental research: sampling, multivariate data analysis and spatial modelling.
In each course there is a significant practical component where participants
are involved in the interpretation of analyses of real environmental data.
Program Directors
Michael Greenacre
Albert Satorra
Target Audience
These graduate level courses are aimed at professionals who are actively
engaged in environmental and ecological research, as well as those involved
in collecting and interpreting environmental data.
Those who will especially benefit from these courses are environmental
scientists, ecologists, regional planners, geographers, agricultural
scientists, meteorologists, geologists, professionals involved in pollution
studies, environmental impact assessment and water resources management,
as well as biologists, botanists and statisticians involved in data
collection and analysis.
Place
The classes will be held at IDEC (Continuing Education Institute, Pompeu
Fabra University) - Balmes, 132. 08008 Barcelona
Language
All the courses will be taught in English
COURSE 1.
SAMPLING PRINCIPLES AND METHODS FOR ENVIRONMENTAL STUDIES
Vic Barnett
The collection of representative data for efficient estimation of
population characteristics is vital in all areas of statistical enquiry -
none more so, however, than in the study of environmental issues. The
methods of survey sampling of finite populations is relevant to environ-
mental sampling and monitoring (e.g., in environmental change networks).
But more modern approaches which depart from the earlier principles of
randomness and segregation can yield major advantages for estimating
elusive environmental characteristics, especially when sample selection
and testing are expensive and optimal procedures are thus vital. The
traditional methods of simple random sampling, stratified sampling and
cluster sampling for finite populations will be presented. They will be
compared and will serve as a background to the discussion of more structured
approaches of composite sampling, ranked set sampling, capture-recapture and
transect sampling, which are specifically tailored for studying environmental
issues.
Vic Barnett is Professor of Environmental Statistics at the University of
Nottingham, England. He has pioneered initiatives in environmental
statistics such as SPRUCE (Statistics in Public Resources, Utilities and
Care of the Environment), with its major international conferences and four
volumes in the series Statistics for the Environment (published by John
Wiley & Co.). He has written and edited many books on statistical research
and methods including the topics of environmental statistics, survey
sampling, outliers and inference.
Date: 26 June, from 9.30 to 13.30 and from 15.00 to 18.00, and 27 June,
from 9.30 to 13.30 hours.
COURSE 2.
MULTIVARIATE ANALYSIS IN ENVIRONMENTAL, ECOLOGICAL AND
ECOTOXICOLOGICAL RESEARCH
Cajo J. F. Ter Braak
Multivariate analysis is a toolbox to efficiently explore and interpret
environmental and ecological data. The usage of the toolbox requires some
basic understanding of the methods, which this course aims to provide.
The emphasis will be on exploring and testing the relationships between
two sets of variables, using the biplot graphic display and Monte Carlo
permutation methods. The two sets of variables could be, for example in
an environmental context, contaminant concentrations versus site or layer
characteristics; or, in an ecological context, species occurrence data
versus environmental variables; or, in an ecotoxicological context, species
data versus experimental conditions. In the last decade, powerful methods
have been developed that relate the two sets directly, in particular with
reduced rank regression (redundancy analysis) and canonical correspondence
analysis. Using real-life examples and the computer program Canoco for
Windows, it will be shown how to choose a strategy of analysis, how to
carry out the required analyses and how to interpret the results.
Special topics will include the analysis of data from Before-After-Control-
Impact experiments and other repeated measurement data such as resulting
from microcosm experiments in ecotoxicology. Elementary knowledge of linear
regression analysis and univariate analysis of variance is assumed.
Cajo J.F. ter Braak is a senior consultant at the Centre for Biometry
Wageningen, the Netherlands, with specialization in multivariate methods
for species-environment relationships. He is co-author of the textbook
Data Analysis in Community and Landscape Ecology (Cambridge University
Press, 1995), and author of the program Canoco, which is the de-facto
standard for ordination of ecological data. Recently, he developed the
Principal Response Curves method for the analysis of microcosm experiments
in ecotoxicology.
Date: 29-30 June, from 9.30 to 13.30, and from 15.00 to 18.00 hours.
COURSE 3.
MODEL-BASED GEOSTATISTICS, WITH ENVIRONMENTAL
APPLICATIONS
Peter J. Diggle
Conventional geostatistical methodology, for example "kriging", provides
optimal solutions to spatial estimation problems only when the spatial
phenomenon being studied can be modelled as a Gaussian stochastic process
and interest is in linear properties of the phenomenon, such as the value
at a point. However, in many applications the Gaussian assumption is
unrealistic and interest is in non-linear properties. For example, in a
study of residual contamination from nuclear weapons testing, data are in
the form of Poisson-distributed counts of photon emissions at various
locations on an island, and interest is in the maximum level of residual
contamination over the island. To deal with problems of this kind, we have
developed a model-based approach to geostatistics, in which we make
explicit assumptions about the probability distributions which generate the
experimental data. Our predictions coincide with those of conventional
geostatistics when we impose Gaussian assumptions and predict linear
properties, but extend the conventional methodology to non-linear and
non-Gaussian problems. Furthermore, our methods make proper allowance for
parameter uncertainty when estimating prediction errors. The course will
introduce the main ideas of model-based geostatistics and demonstrate
their application to environmental data. Course topics include: a model-
based derivation of ordinary kriging, estimating spatial covariance
structure, extending the model-based approach to non-linear, non-Gaussian
problems, and various case studies in the environmental sciences. This
course assumes a basic knowledge of probability theory and regression
modelling, as well as familiarity with elementary concepts in time series
analysis and stochastic processes such as the autocorrelation function.
Course notes will include a summary of the necessary background.
Peter Diggle is Professor of Statistics at Lancaster University, U.K.
He is the author of four books and about 90 articles, many of these being
in the field of spatial statistics. In 1997 he was awarded the Royal
Statistical Society Guy Medal in Silver.
Date: 1-2 July, from 9.30 to 13.30, and from 15.00 to 18.00 hours.
Course fees
The course fees include course notes, a Diploma of Institut d'Educacio
Continua, meals and refreshments during the courses.
Course 1: 60.000 ptas (40.000 ptas. if paid before the 20th of May 1998).
Course 2: 75.000 ptas (50.000 ptas. if paid before the 20th of May 1998).
Course 3: 75.000 ptas (50.000 ptas. if paid before the 20th of May 1998).
For those who register for 2 courses the fee is reduced by 10%.
For those who register for 3 courses the fee is reduced by 20%.
SPECIAL FEES for doctoral students and participants in a previous APPLIED
STATISTICS WEEK:
Course 1: 30.000 ptas.
Course 2: 37.500 ptas.
Course 3: 37.500 ptas.
Registration
The registration form must be sent by the 20th of May. Payment can be made
by bank transfer (including a copy of the bank transfer), by credit card
(VISA) or by bank cheque to the Institut d'Educacio Continua. In the case
of foreign payments, the cheque should be in convertible pesetas.
Accommodation facilities
The Continuing Education Institute has arranged a special price of 9.500 pts.
with some hotels in Barcelona for the accommodation of participants:
Astoria ***, address: Paris, 203 (300m from IDEC)
Balmes ***, address: Mallorca, 216 (200m from IDEC)
In all cases, prices are for a double room for either one or two persons
(breakfast and VAT included).
Information and enrolment
IDEC (Continuing Education Institute)
Balmes, 132
08008 Barcelona - Spain
Telephone: (34-3) 542 18 06/15/00
Fax: (34-3) 542 18 08
http://www.upf.es/idec
E-Mail: [log in to unmask]
APPLIED STATISTICS WEEK '98
R E G I S T R A T I O N F O R M
Send to: Registration - IDEC
Balmes, 132. 08008 Barcelona - Spain
Fax Number: (34-3) 542 18 08
From:
Name...................Surname..........................................
Mailing Address ........................................................
........................................................................
Telephone ..................Fax....................
E-mail.............................................
Organization............................................................
Position................................................................
REGISTRATION (Mark your choice with an X)
[ ] Sampling Principles and Methods for Environmental Studies
[ ] Multivariate Analysis in Environmental, Ecological and
Ecotoxicological Research
[ ] Model-Based Geostatistics, with Environmental Applications
ACCOMMODATION
[ ] Hotel Balmes *** [ ] Hotel Astoria ***
Number of nights ......... From ............. to ..............
CHARGES (Please calculate the total amount)
Course Fees + .................
(According to the date of registration
or the doctoral special fees)
Reduction - .................
(10% or 20% if you apply for 2 or 3 courses.
This reduction is not available with special fees)
Total course fees: …..............
Hotel + .................
(9.500 pts x number of nights)
Final total: .................
PAYMENT (Mark your choice with an X)
[ ] Bank Cheque to Institut d'Educacio Continua
[ ] Bank Tranfer to 0015-4444-11-0100032922 Banca Catalana
(a copy of the bank transfer should be sent)
[ ] Credit Card Payment
Visa card number: ..........................................
Expiry Date:........................
Cardholder's printed name: .................................
Date: Signature:
________________________________________________________________________
--------------------------------------------------------
| Jan Graffelman | internet: [log in to unmask] |
| Universitat Pompeu Fabra | tel: 34-3-5421621 |
| Ramon Trias Fargas 25-27 | fax: 34-3-5421746 |
| 08005 Barcelona, Spain | |
--------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
