Remove

[Craigloehl]

 
In a message dated 11/4/2009 6:10:14 P.M. Central Standard Time,  
LISTSERV@... writes:

There  are 17 messages totalling 1009 lines in this issue.

Topics of the  day:

1. Lecture day at Durham, Prof. Balakrishnan visiting the  UK
2. ANNOUNCE - Int J of Molecular Epidemiologyy &  Genetics
3. Second UK One-Day Meeting on Morphometrics and  Statistical Shape
Analysis
4. COURSE: Stata & Matlab courses at  Imperial College
5. SHORT COURSE in Factor Analysis & Structural  Equation Modelling
6. Power analysis book for multi-level  regressions and other advanced
methods
7. MASTER CLASS: GAMLSS and  P-Splines In Action.
8. Fwd: Principal Components Analysis (PCA) vs  Factor Analysis using SPSS
9. David Cox's Talk @ UL
10. Standard  Reference Period
11. SEMINARS: RSS Leeds/Bradford Local Group  meeting
12. David Cox's Talk - website fix
13. Repeated measurement in  categorical data (2)
14. 3rd CFP Agent-Directed Simulation, April 12-14,  2010, Orlando, Florida
15. EPSRC CASE studentship - EDINBURGH
16. RSS  Medical Section Meeting: 30th November   Communicating the Risks  
to
Health - Whose Responsibility is  it?

----------------------------------------------------------------------

Date:   Wed, 4 Nov 2009 09:22:30 +0000
From:    Jochen Einbeck  <jochen.einbeck@...>
Subject: Lecture day at Durham, Prof.  Balakrishnan visiting the UK

Prof N. Balakrishnan (McMaster University,  Canada) will be visiting
Durham University from 9 to 16 April  2010.

Prof Balakrishnan has been actively involved in research in many  areas of=

Statistics, and has made significant contributions to, among  other topics=
,
Models and Analysis of Medical and Lifetime Data,  Life-Testing and Reliab=
ility,
Order Statistics, Robust Inference,  (Multivariate) Distribution Theory,
Characterization Theory, Inferential  Methods, Industrial Statistics,
Nonparametric Inference, Outliers,  Multivariate Analysis,
Bayesian and Empirical Bayesian Inference,  Combinatorial Applications to
Probability and Statistics, Record Values and  Processes, Theory of Runs
and Scans, Waiting Time Problems, Ranked Set  Sampling, and Statistics in
Finance.

He is currently   Editor-in-Chief for Communications in Statistics -
Theory and Methods and  for Communications in Statistics - Simulation and
Computation, and  Executive Editor for Journal of Statistical Planning and=

Inference. He  is Editor-in-Chief for the Revised Edition of Wiley's Encyc=
lopedia
of  Statistical Sciences. For more details  see:

http://www.math.mcmaster.ca/bala/bala.html

On Wednesday 14  April, he will give a 2-hour lecture, as part of a
lecture day organised in  Durham - see below for further details.
In addition, he will give the  following two research seminars:

Monday 12 April, Edinburgh  University:
`Over- and under-dispersed Poisson distributions and  processes'
for details contact Natalia Bochkina  (n.bochkina@...)

Thursday 15 April, Newcastle University:
'On  some stochastic orderings and related characterizations for some
discrete  and continuous distributions'
for details contact Jordan Stoyanov  (jordan.stoyanov@...)

-------------------------------------------------------------------------=
-
Wednesday  14 April, Durham University:=20=20

LECTURE DAY, with the following  speakers and topics:

1. Narayanaswamy Balakrishnan (McMaster  Univ)=20
`Permanents, order statistics, outliers and robustness'

2.  Chris Jones (Open Univ)
`The Cauchy-Schlomilch transformation, its  extensions, and a useful analo=
gue'=20

3. Jordan Stoyanov  (Newcastle Univ)
'Non-linear transformations of random data: moment  determinacy of their
distributions'

4. Jochen Einbeck (Durham  Univ)=20
`Data compression and regression based on local principal curves  and mani=
folds'

5. Tahani Maturi (Durham Univ)
`Nonparametric  predictive inference for comparison of lifetime data'


Further  details about the lecture day are available  on:

http://www.maths.dur.ac.uk/~dma0je/bala/

Everybody is most  welcome to attend this lecture day - if you wish to do
so please read the  details on the webpage and contact us with the  informa=
tion
requested.

------------------------------

Date:   Wed, 4 Nov 2009 10:00:35 -0000
From:    Mario Cortina  <m.cortina@...>
Subject: ANNOUNCE - Int J of Molecular  Epidemiologyy & Genetics

The International Journal of Molecular  Epidemiology and Genetics (IJMEG)
(ISSN 1948-1756), is an open
access,  online only journal to facilitate rapid dissemination of novel
discoveries  and new methodologies in
medical molecular epidemiology and  genetics.  It was founded by a group of
pre-clinical and clinical  
academic researchers from around the world, who are devoted to the  
promotion
and advancement of our
understanding of the molecular basis  of disease.

The scope of IJMEG is intended to encompass that of  multi-disciplinary
researchers from any scientific
discipline where the  primary focus of the research is to increase and
integrate knowledge about  chronic
disease aetiology with the ultimate aim of advancing the cure and  
prevention
of chronic disease and the
overall improvement of public  health. To achieve these aims IJMEG
encourages
research that  uses
molecular, cellular and biochemical concepts and techniques, including  the
wealth of information and
application that biomarkers can bring,  that either complement current
epidemiological knowledge or
which are  directly incorporated into well-designed epidemiologic or  
genetic
studies.  

Unlike most other open access online  journals, IJMEG will keep most of the
traditional features of paper print  
that we are all familiar with, such as a unique cover page for every  single
issue, continuous volume and issue
numbers, as well as  continuous page numbers to retain our comfortable
familiarity towards an  academic
journal.

The first issue of IJMEG will be officially  launched in January, 2010 and
published as a quarterly journal. We
are  accepting manuscript submissions at this time- see details on  IJMEG
<http://www.ijmeg.org/files/IJMEG_CALL_FOR_PAPERS.pdf> CALL FOR  PAPERS.
All
manuscripts will be peer reviewed and published within two  weeks if
accepted. Please e-mail the manuscript
to the   <mailto:editorial@...> Editorial Office if you are  
interested
in publishing your work in IJMEG. If you are interested in  submitting
a manuscript to IJCEM for consideration of publication, please  read the
journal's  <http://www.ijmeg.org/ForAuthors.html>  Guidelines for Author
and
visit   <http://www.ijmeg.org/Submission.html>  submission.







Dr Mario Cortina Borja; Senior  Lecturer in Statistics

Centre for Paediatric Epidemiology and  Biostatistics

Institute of Child Health, University College  London

30 Guilford Street London, WC1N 1EH, UK

Tel:  [phone]; FAX:  [phone];

M.Cortina@...

http://www.ucl.ac.uk/~sejjmco



------------------------------

Date:   Wed, 4 Nov 2009 10:10:17 +0000
From:    "A.Kume"  <A.Kume@...>
Subject: Second UK One-Day Meeting on  Morphometrics and Statistical Shape
Analysis

Dear  Colleagues

This is to announce the second One-Day Meeting on  Morphometrics and
Statist=
ical Shape Analysis.=20

It will take  place on 11th January 2010 at University of Kent, UK.

The full  programme an registration details are on:  
http://www.flywings.org.=
uk/MorphoMeet10/index.htm.

The  organizers=

------------------------------

Date:     Wed, 4 Nov 2009 11:12:12 +0000
From:    "Kells, Richard"  <r.kells@...>
Subject: COURSE: Stata & Matlab courses  at Imperial College

DATA MANAGEMENT & STATISTICAL ANALYSIS USING  STATA at Imperial College
Lond=
on, during November  2009.

INTRODUCTION TO PROGRAMMING USING MATLAB at Imperial College  London,
during=
December 2009.

Data Management & Statistical  Analysis Using  
Stata</stathelp/courses/statac=
ourses/datamanagementandstatisticanalysisusingstata>
Wed  25 November & Wed 02 December (10 - 5pm)

Cost:  =A3390.

Introduction to Programming Using Matlab
Wed 09 & Wed 16  December (2 - 5pm)

Cost: =A3195.

**Please note that we also  provide courses in SPSS, Stata, Matlab,
Clinical=
Trials & R. Please  contact 0207 594 3856 or  
stathelp@...<mailto=
:stathelp@...> for further  information**

**We also provide STATISTICAL ADVICE on a one-to-one  consultation basis.
Pl=
ease contact  stathelp@...<mailto:stathelp@...> , or
v=
isit  http://www3.imperial.ac.uk/stathelp for further details**

For further  information please contact Richard on 0207 594 3856 or  
stathelp=
@imperial.ac.uk<mailto:stathelp@...>  .
Alternatively an application form and accompanying course outline can be  
do=
wnloaded from our web-site:  http://www3.imperial.ac.uk/stathelp

------------------------------

Date:   Wed, 4 Nov 2009 11:15:59 +0000
From:    Bianca L De  Stavola <Bianca.DeStavola@...>
Subject: SHORT COURSE in  Factor Analysis & Structural Equation  Modelling

________________________________________________
Factor  Analysis & Structural Equation Modelling:=20
An Introduction Using  Stata and Mplus

17-19 February 2010

London School of Hygiene  & Tropical  Medicine
________________________________________________


COURSE  ORGANIZERS
George Ploubidis and Bianca De Stavola

THE  COURSE

Most courses on factor analysis and structural equation  modelling
concentrate on the use of traditional factor analytic models  with
interval level data and neglect the latest developments within  the
Generalised Latent Variable Modelling framework. The course  will
discuss
the current state of the art with respect to factor  analysis and
structural equation modelling, while retaining a practical  focus.
Participants will acquire awareness of the new available methods  and
gain competence in applying and combining these in simple  settings.

Aims:
Understand measurement principles such as  reliability and=20
precision Perform Exploratory Factor Analysis and  Confirmatory Factor
Analysis
with a mixture of binary/ordinal/interval  data.=20
Estimate and interpret path analytic and structural models  =20
Estimate and interpret growth curve and latent class models
Develop  awareness of the common threads across these methods
Gain practical  experience in using these methods using Stata and Mplus


COURSE FEE:  =C2=A3 600

FOR FURTHER INFORMATION=20
contact:  george.ploubidis@...=20
or visit:   http://ww.lshtm.ac.uk/prospectus/short/sfasem.htm=20

TO  REGISTER
contact:  shortcourses@...=20


(apologies for multiple  posting)

------------------------------

Date:    Wed,  4 Nov 2009 12:15:42 +0000
From:    Evangelos Kontopantelis  <E.Kontopantelis@...>
Subject: Power analysis book for  multi-level regressions and other
advanced methods

Dear  all

=


Can anyone suggest a good book that covers power  analysis for complex
stati=
stical methods?

Please use  e.kontopantelis@... and I'll provide a summary for the
mai=
ling  list.

=


Thank you in  advance

=


Evan

=


Evangelos Kontopantelis,  PhD
Research Associate in Statistics
NPCRDC, Williamson building 5th  floor
University of Manchester
M13  9PL

------------------------------

Date:    Wed, 4 Nov  2009 14:20:36 -0000
From:    "Stewart, Deborah"  <d.l.stewart@...>
Subject: MASTER CLASS: GAMLSS and  P-Splines In Action.

APOLOGIES FOR CROSS POSTING.

****Please  would you kindly circulate this to all academic staff, =
postgraduate  research students and MSc students within your school or  =
department****

INTRODUCTION TO MODERN SMOOTHING METHODS:=A0 GAMLSS  AND P-SPLINES IN =
ACTION

3 DAY MASTER CLASS AT THE POSTGRADUATE  STATISTICS CENTRE, LANCASTER =
UNIVERSITY ON 30 NOVEMBER - 2 DECEMBER  2009

PRESENTED BY MIKIS STASINOPOULOS, PAUL EILERS AND ROBERT  RIGBY.

This short course introduces two powerful ideas in modern  statistical =
modelling, Generalized Additive Models for Location, Scale  and Shape =
(GAMLSS) and the P-Splines approach to  smoothing.=A0=20

GAMLSS were introduced by Rigby and Stasinopoulos  (2005) as a way of =
overcoming some of the limitations associated with the  popular =
Generalized Linear Models (GLM) and Generalized Additive Models  =
(GAM).=A0 This framework generalises methodologies such as fractional  =
polynomials and the LMS method for constructing reference ranges and  =
normalised growth centile standards. For centile estimation, the WHO  =
Multicentre Growth Reference Study Group have recommended GAMLSS and the  =
Box-Cox Power Exponential distributions for the construction of the WHO  =
Child Growth Standards. GAMLSS is implemented in a series of R  =
packages.=A0 P-Splines (Eilers and Marx, 1996) is a flexible tool for  =
smoothing, based on penalised regression techniques. Its simplicity and  =
flexibility allows its use in a variety of different practical  =
applications.=20

This three day course will use the first day as  an introduction to =
GAMLSS, the second day as an introduction to P-Splines  and the third day =
for further applications and how the two approaches can  be combined. =
Each lecture session will be followed by a practical session  in R.


Cost:=A0 Postgraduate Students - =A390 / Academic Staff from  HE =
Institutions - =A3180 / Participants from non-HE Institutions -  =A3660

For booking registration and further details please go to:=A0  =
http://psc.maths.lancs.ac.uk/shortCourses/?q=3Dnode/48


Deborah  Stewart
Postgraduate Statistics Centre Secretary
Dept of Mathematics  & Statistics
Lancaster University
Lancaster
LA1  4YF

Tel:=A0 [phone]
Fax: [phone]
Email:  d.l.stewart@...
http://www.maths.lancs.ac.uk/psc

------------------------------

Date:   Wed, 4 Nov 2009 14:28:20 +0000
From:    Roman Ortuno  <romeror@...>
Subject: Fwd: Principal Components Analysis (PCA) vs  Factor Analysis using
SPSS

Dear All,

Could someone explain me  the practical difference between Principal
Compoents Analyisis (PCA) vs.  Factor Analyisis (FA) when using SPSS (16.0)?
Does the difference have to  do with the extraction method used? (i.e.
Principal components, vs.  others?).

Also, what are the dangers of including dichotomous variables  in PCA/FA and
are there any examples published studies where this was  done?

Many thanks in advance for your help,

Roman  Romero-Ortuno
PhD student,  TCD

------------------------------

Date:    Wed, 4 Nov  2009 14:53:28 +0000
From:    "gilbert.mackenzie"  <gilbert.mackenzie@...>
Subject: David Cox's Talk @ UL

Dear  All

The mp3 voice files covering the talk are available  at
www.il.ie/biostatistics and also please see the new
discussion page  by following the blue wedge from
seminars.

Best for  now

Gilbert

--  



_____________________________

Prof. Gilbert  MacKenzie
Centre of Biostatistics,
Room B2034
Dept. of Mathematics  & Statistics,
University of Limerick,
Limerick  
Ireland

CBS ~ http://www.ul.ie/biostatistics

BIO-SI ~  http://www.ul.ie/bio-si

Gilbert ~  http://www.staff.ul.ie/mackenzieg

Email:  gilbert.mackenzie@...

Tel: [phone]
Fax: [phone]

ISA ~  http://www.istat.ie.
_________________________

------------------------------

Date:   Wed, 4 Nov 2009 15:12:39 -0000
From:    "Allan Reese  (Cefas)" <allan.reese@...>
Subject: Standard Reference  Period

I frequently see references to the "standard reference period"  for
weather data and it seems generally to be quoted (eg by Philip Eden)  as
1971-2000.  On looking for a source, I found however  that,

"The 30-year period 1961 to 1990 has been designated as  the
international standard reference period for climate averages by  the
World Meteorological Organization. Averages for the period 1971 to  2000
have also been produced for the UK, but the earlier period has  been
chosen for the UKCIP grids as it represents a better baseline  for
placing recent climate change into  context."
http://www.metoffice.gov.uk/climatechange/science/monitoring/ukcp09/faq.
html#faq

The  WMO site states "Under the Technical Regulations (WMO No.  49),
climatological standard normals are averages of climatological  data
computed for the following consecutive periods of 30 years: 1  January
1901 to 31 December 1930, 1 January 1931 to 31 December 1960, and  so
forth. The most recent period for climatological standard normals  is
from 1961 to 1990, and the next period for the calculation  of
climatological standard normals will be 1991 to  2020."
http://www.wmo.int/pages/prog/gcos/documents/gruanmanuals/WCP_CCl/guide_
third_edition_draft_may2007.pdf

What  statistical fiddle is being perpetrated by having the UK reference
period  roll forward a  decade?

Allan=20


***************************************************************************=
********
This  email and any attachments are intended for the named recipient only.  
 =
Its unauthorised use, distribution, disclosure, storage or copying is not  
p=
ermitted.  If you have received it in error, please destroy all  copies and
=
notify the sender.  In messages of a non-business nature,  the views and
opi=
nions expressed are the author's own and do not  necessarily reflect those
o=
f the organisation from which it is  sent.  All emails may be subject to  
mon=
itoring.
***************************************************************************=
********

------------------------------

Date:   Wed, 4 Nov 2009 15:54:42 +0000
From:    Paul Baxter  <p.d.baxter@...>
Subject: SEMINARS: RSS Leeds/Bradford Local  Group meeting

Dear all,

On Wednesday 18th November, the RSS  Leeds/Bradford local group will be
hosting an afternoon on  "statistics
and the credit crunch" featuring talks by Mick Ellender  (Callcredit
Ltd), David Hand (RSS President), Nick Bingham (Imperial  College) and
Klaus Schenk-Hoppe (University of Leeds).

Further  details can be found on our  webpage:

http://tinyurl.com/rss-lba

All welcome!

Regards,  Paul
===================================================================
Dr.  Paul D. Baxter
Secretary/Treasurer, RSS Leeds/Bradford Local  Group,
Division of Biostatistics, University of Leeds, Leeds, LS2 9JT,  UK.
-------------------------------------------------------------------
Leeds/Bradford:  Wednesday 18 November, 2.00pm, Leeds University.

Statistics and the  credit crunch

Mick Ellender (Callcredit Ltd) [2pm -  2.40pm].

Credit scoring in the current economic climate

A high  level look at the statistical techniques and methodologies used
in credit  scoring and its use through the consumer life cycle from point
of  acquisition to collections and recoveries. The high level theories
will be  reviewed along with practical case studies and also a look at
the  challenges facing lenders in the current economic environment.

David  Hand (Royal Statistical Society) [2.50pm - 3.30pm].

Statistics, data  mining, and the personal credit scoring industry

The modern consumer  credit industry is fundamentally data-driven,
employing statisticians to  build models for an increasing variety of
problems. This talk describes  some of these problems, and the sorts of
statistical models which have been  built to tackle them.

Nick Bingham (Imperial College) [4pm -  4.40pm].

Title TBA

Klaus Schenk-Hoppe (University of Leeds)  [4.50pm - 5.30pm].

Going naked: The effect of short-selling  bans

Short-selling of shares is a popular way of betting your beliefs  as a
"bearish" investor. In the current financial crises, short-selling  has
been blamed for the unprecedented decline in prices and for putting  the
entire financial system at risk. This talk will give a  non-specialist
introduction to the topic and explain the statistical and  modelling
issues in quantifying the effect of short-selling  bans.

The meeting will be held at Leeds University Roger Stevens  Building in
RSLT 24, starting at 2.00pm. Refreshments will be available  from 1.30pm
in the foyer of level 9 of the School of Mathematics. A  further
refreshment break will be held between  3.30pm-4pm.

------------------------------

Date:     Wed, 4 Nov 2009 15:56:15 +0000
From:    "gilbert.mackenzie"  <gilbert.mackenzie@...>
Subject: David Cox's Talk - website  fix

Dear All

The mp3 voice files covering the talk are available  at
www.ul.ie/biostatistics and also please see the new
discussion page  by following the blue wedge from
seminars.

Best for  now

Gilbert

--  



_____________________________

Prof. Gilbert  MacKenzie
Centre of Biostatistics,
Room B2034
Dept. of Mathematics  & Statistics,
University of Limerick,
Limerick  
Ireland

CBS ~ http://www.ul.ie/biostatistics

BIO-SI ~  http://www.ul.ie/bio-si

Gilbert ~  http://www.staff.ul.ie/mackenzieg

Email:  gilbert.mackenzie@...

Tel: [phone]
Fax: [phone]

ISA ~  http://www.istat.ie.
_________________________

------------------------------

Date:   Wed, 4 Nov 2009 08:53:25 -0800
From:    PGW  <pgw888@...>
Subject: Repeated measurement in categorical  data

Hello, =0A=0AI have a set of data with 3 repeated measurements in  
categoric=
al data 1, 2, 3, or 4.  and like to see if there is any  increase along the
=
time.  I would use ANOVA if it was a continuous  data.  =0A=0AIt would be
ve=
ry much appreciated if anyone could give  me some advice on how to tackle
ca=
tegorical data with repeated  measurement.=0A=0AThank you very much in
advan=
ce.=0A=0AKind  regards=0A=0AJenny=0A=0A=0A      

------------------------------

Date:    Wed, 4 Nov  2009 11:22:50 -0600
From:    Yu Zhang  <yzhang@...>
Subject: 3rd CFP Agent-Directed Simulation,  April 12-14, 2010, Orlando,
Florida

This message is in MIME  format.  The first part should be readable text,
while the  remaining parts are likely unreadable without MIME-aware  tools.

---1031597753-1709324360-1240492342=:21248
Content-Type:  TEXT/PLAIN; CHARSET=UTF-8; FORMAT=flowed
Content-ID:  <Pine.LNX.4.64.0905010805222.8515@...>
Content-Transfer-Encoding:  quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by  
bofur.jiscmail.ac.uk id  nA4Hl3vk015465

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
CALL  FOR PAPERS and POSTERS
Agent-Directed Simulation Symposium  (ADS'10)
Orlando, Florida, USA
April 12-14,  2010

http://www.site.uottawa.ca/~oren/conf-org/ADS_2010/ADS-CFP.htm
Manuscript  Submission: November 30, 2009.

Sponsored by The Society for Modeling  and Simulation
International (SCS) in collaboration with  ACM/SIGSIM.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

As  part of the 2010 Spring Simulation Multi-conference  (SpringSim'10)=20
http://www.scs.org/confernc/springsim/springsim10/springsim10.htm

the  2010 Agent-Directed Simulation Symposium is a premier platform to
explore  all three aspects of the synergy of simulation and agent
technologies.  Hence, it has a special place within simulation and
agent conferences,  including agent-based (social) simulation
conferences. Therefore the ADS  symposium fills a gap in the agent
community as well as the simulation  community.

The purpose of the ADS symposium is to facilitate  dissemination of the
most recent advancements in the theory, methodology,  application, and
toolkits of agent-directed simulation. Agent-directed  simulation is
comprehensive in the integration of agent and simulation  technologies,
by including models that use agents to develop  domain-specific
simulations, i.e., agent simulation (this is often referred  to as
agent-based simulation -when other two important aspects are  not
considered), and by also including the use of agent technology  to
develop simulation techniques and toolkits that are  subsequently
applied, either with or without agents.

Hence,  agent-directed simulation consists of three distinct, yet
related areas  that can be grouped under two categories as follows:

1.  Simulation for Agents (agent simulation): simulation of agent
systems in engineering, human and social dynamics, military
applications etc.
2. Agents for Simulation  (which has two aspects): agent-supported
simulation  deals with the use of agents as a support facility to
enable computer assistance in problem solving or enhancing
cognitive capabilities; and agent-based simulation that  focuses
on the use of agents for the generation of  model behavior in a
simulation study.

Through  the theme of agent-directed simulation, the symposium will
bring together  agent technologies, tools, toolkits, platforms,
languages, methodologies,  and applications in a pragmatic manner. In
this symposium, established  researchers, educators, and students are
encouraged to come together and  discuss the benefits of agent
technology in their use and application for  simulation. It is a way
for people to discuss why and how they have used  agent technology in
their simulations, and describe the benefit of having  done so.


The theme of ADS'10 is based on the observation of the  following
premises.

* The growth of new advanced  distributed computing standards along
with the rapid  rise of e-commerce are providing a new context that
acts as a critical driver for the development of next generation
systems. These standards revolve around  service-oriented
technologies, pervasive computing,  web-services, Grid, autonomic
computing, ambient  intelligence etc. The supporting role that
intelligent  agents play in the development of such systems is
becoming pervasive, and simulation plays a critical role in the
analysis and design of such systems.

* The  use of emergent agent technologies at the organization,
interaction (e.g., coordination, negotiation, communication) and
agent levels (i.e. reasoning, autonomy) are expected to  advance
the state of the art in various application  technologies is
difficult. Using agent-supported  simulation techniques for
testing complex agent systems  is up and coming field.

* To facilitate bridging the gap  between research and application,
there is a need for  tools, agent programming languages, and
methodologies  to analyze, design, and implement complex,
non-trivial  agent-based simulations. Existing agent-based
simulation tools are still not mature enough to enable developing
agents with varying degrees cognitive and reasoning  capabilities.

ADS'10 will provide a leading forum to bring together  researchers and
practitioners from diverse simulation societies within  computer science,
social sciences, engineering, business, education, human  factors, and
systems engineering. The involvement of various  agent-directed
simulation groups will enable the cross-fertilization of  ideas and
development of new perspectives by fostering novel advanced  solutions,
as well as enabling technologies for agent-directed  simulation

AUTHOR GUIDE

* Technical papers provide  a longer format for presenting experience
reports,  research results, or descriptions of  "work in progress".
They are limited to 8 pages.

* Short position  papers are targeted at raising a question or framing
an  issue for discussion during the symposium. Position papers are
limited to 3 pages.

* Poster presentations  present an opportunity to present work in
progress and  receive feedback from colleagues. A one page write-up
of the poster presentations will be included in the proceedings.

(For  all, required font sizes are: min 10 pt for text and min 9 pt for
figures  or references).

Papers should be submitted electronically to
http://www.softconf.com/scs/ADS10/ .
All papers will be  subject to a peer-reviewing process by three program
committee members.  (Please see the key dates listed below.)


FINAL PAPER SUBMISSION  GUIDELINES

All prospective authors, whose papers are accepted for  inclusion in the
program, will be invited to submit their position or  technical  papers
to ADS'10. Accepted and registered papers will be  published in the
conference proceedings by the SCS. In addition, the  committee will
select a set of best papers. Authors of these papers will be  encouraged t=
o=20
submit appropriately expanded versions of these  papers for journal
publication.


KEY DATES
Nov  30, 2009: Manuscript submission
Dec 30, 2009: Notification of  acceptance
Jan 20, 2010: Full Camera-ready  papers
Apr 12-14, 2010: ADS'10 Symposium


General  Co-Chair
Levent Yilmaz, Auburn University
Tuncer =C3=96ren, University of Ottawa
Program Co-Chair
Gregory Madey, University of Notre Dame
Maarten Sierhuis,  Carnegie Mellon University, NASA Ames Research Cente=
r
Yu  Zhang, Trinity  University
---1031597753-1709324360-1240492342=:21248--

------------------------------

Date:   Wed, 4 Nov 2009 18:05:38 +0000
From:    Colin Aitken  <C.G.G.Aitken@...>
Subject: EPSRC CASE studentship -  EDINBURGH

UNIVERSITY OF EDINBURGH

MASS SPEC ANALYTICAL (MSA),  BRISTOL.

EPSRC CASE studentship.

Establishment of frameworks for  the evaluation of evidence relating to=20
traces of drugs

Joseph  Bell Centre for Forensic Statistics and Legal Reasoning, School=20
of  Mathematics, The University of Edinburgh,

Mass Spec Analytical (MSA),  Bristol.


A PhD CASE studentship is available at the  Joseph Bell Centre for=20
Forensic Statistics and Legal Reasoning in the  School of Mathematics in=20
collaboration with Mass Spec Analytical (MSA),  Filton, Bristol,=20
(http://www.msaltd.co.uk) under the supervision of  Professor Colin Aitken.

Fees and stipend are available  for UK nationals. The stipend will be=20
the current EPSRC rates with an  enhancement from the industrial partner=20
and is expected to be about  =A316,000 in the first year. There is also a=20
travel budget to enable  attendance at conferences and for research visits.

Further particulars, including more details of the project, are
available  on the School of Mathematics  website:

http://www.maths.ed.ac.uk/local/jobs/vacancies/

The successful candidate will be based in the School of
Mathematics  at the University of Edinburgh, with excellent library
and computing  facilities and a flourishing postgraduate community.

It  is part of the condition of the award that the student spends
at least  three months of the project working at MSA in Bristol. It
is expected this  will be in three annual periods of one month
each. MSA will provide the  cost of travel to and from Bristol and
of accommodation for this  placement.

Qualifications:

Applications are  invited from graduates with a first-class Honours or=20
Master's degree in  a mathematics degree with a substantial statistics=20
component or a  statistics degree or from final-year undergraduate or=20
Masters students  on such a degree programme, or equivalent.

Vetting:  Applicants  may be subject to UK security vetting.

Start date:

This is open to negotiation but the latest possible
starting date is  October 1st 2010.

Application:

Please submit a CV, the names and addresses of two
referees and a covering  letter to Ms Gill Law, Graduate School
Secretary, School of Mathematics,  King's Buildings, The University
of Edinburgh, Mayfield Road, Edinburgh,  EH9 3JZ; g.law@...;
fax [phone].  Electronic applications  are acceptable.

Closing date:  November 30th  2009.

Enquiries are welcome and can be made to Professor Colin  Aitken
(c.g.g.aitken@...).
--=20
Professor Colin  Aitken,
Professor of Forensic Statistics,
School of Mathematics,  King=92s Buildings, University of Edinburgh,
Mayfield Road, Edinburgh, EH9  3JZ.

Tel:    [phone]
E-mail:   c.g.g.aitken@...
Fax :  [phone]
http://www.maths.ed.ac.uk/~cgga


The University of Edinburgh  is a charitable body, registered in
Scotland, with registration number  SC005336.

------------------------------

Date:    Wed,  4 Nov 2009 19:17:30 +0100
From:    Francis Levira  <levira17@...>
Subject: Re: Repeated measurement in categorical  data

Dear Jenny,

There is an extension of logistic regression  that account for the
categorical data with more than two category which is  called
*Proportional Odds Logistic regression*. Using this model the  probability
of
observing a lower response  versus higher one is  modeled.
Since you have repeated measure data, a modified GEE (Generalize  Estimating
Equation) version of analyzing longitudinal data will be used.  The
important
thing is to have correct assumption about the correlation  structure between
your repeated measurement. Marginal or Subject specific  model can be
fitted.

Hope this will give you a light during your  analysis.

Francis.




On Wed, Nov 4, 2009 at 5:53 PM,  PGW <pgw888@...> wrote:

> Hello,
>
> I have  a set of data with 3 repeated measurements in categorical data 1,
2,
>  3, or 4.  and like to see if there is any increase along the time.   I
would

> use ANOVA if it was a continuous data.
>
> It  would be very much appreciated if anyone could give me some advice on
>  how to tackle categorical data with repeated measurement.
>
>  Thank you very much in advance.
>
> Kind regards
>
>  Jenny
>
>
>
>

--
Center for  Statistics
Hasselt University
Patersstraat  15
Diepenbeek
Belgium.
Mobile:[phone]

------------------------------

Date:   Wed, 4 Nov 2009 19:14:01 +0000
From:    Dr Philip Sedgwick  <sgju511@...>
Subject: RSS Medical Section Meeting: 30th  November   Communicating the
Risks to Health - Whose Responsibility  is it?

Apologies for Cross Postings

Royal Statistical Society,  Medical Section Meeting

Communicating the Risks to Health - whose  responsibility is it?
Monday 30th November 2009 2.00 to 5.00pm.
Royal  Statistical Society, 12 Errol Street, London, EC1Y 8LX.
Nearest underground  stations: Barbican, Old Street , Moorgate and=20
Liverpool Street.
see  http://www.rss.org.uk/main.asp?page=3D1759

Attendence is free but it is  recommended that you register if interested=20
in attending. Go to the  following  page:
http://www.rss.org.uk/main.asp?group=3D&page=3D1321&event=3D756&month=3D1=
1&year=3D2009&date=3D=20


Alternatively  visit www.rss.org.uk, then select Meetings and Events,=20
Events Calendar  and then November; scroll down page to select above=20
meeting. A link for  registering is provided.

The meeting will discuss the challenges and  issues when communicating=20
the risks to health to members of the public.  Whose responsibility is it=20
to communicate these risks so they are  clearly understood by members of=20
the public, enabling informed decision  making.

Speakers and titles:

Professor Jane Hutton. Department  of Statistics, University of Warwick.
Communicating risk: duties, virtues  and utilities.

Professor David Spiegelhalter. Winton Professor of the  Public=20
Understanding of Risk, Cambridge University.
It's the  statistician's fault if risks are misunderstood: discuss.

Nigel Hawkes.  Director of Straight Statistics.
Risk communication in the media: friend or  foe?

Nicole Crosby-McKenna Patient group representatives for Women  at=20
Epilepsy Action.

Janine Winterbottom (Nurse Specialist, The  Walton Centre NHS Foundation=20
Trust)
Pregnancy risk and  decision-making: the experience of women with=20
epilepsy preparing for  pregnancy.

For further details please contact meeting organiser: Philip  Sedgwick.=20
St. George=92s, University of London.  (p.sedgwick@...).
There will be time for specific questions after  each talk and towards=20
end of meeting for an open floor  discussion.

------------------------------

End of allstat Digest  - 3 Nov 2009 to 4 Nov 2009  (#2009-274)
************************************************************