Is your email address right? I have a note for you.J.Nelder.
-----Original Message-----
From: A UK-based worldwide e-mail broadcast system mailing list on behalf of Automatic digest processor
Sent: Thu 10/21/2004 12:01 AM
To: Recipients of allstat digests
Cc:
Subject: allstat Digest - 19 Oct 2004 to 20 Oct 2004 (#2004-245)
There are 7 messages totalling 220 lines in this issue.
Topics of the day:
1. Statistician contract in Continental Europe
2. relative risk or odds ratio
3. Short Course in "R"
4. Question on Clustering (2)
5. REMINDER: RSS HIGHLANDS GROUP MEETING, 21/10/2004: John Norrie
6. Medical stat
----------------------------------------------------------------------
Date: Wed, 20 Oct 2004 09:22:27 +0100
From: Mick Messing <[log in to unmask]>
Subject: Statistician contract in Continental Europe
Dear Allstaters
My client is looking for an experienced Statistician to work in a
Biostatistics department, which plays a key role in successful drug
development by operating on an international level and contributing expert
input to the drug development processes.
Working closely with other statisticians and colleagues in drug development,
especially data management and clinical research, biostatistics contributes
to clinical trial activities within global projects.
You will be responsible for the analysis and reporting of international
clinical trials for regulatory submissions.
The candidate must have a university degree in statistics or related subject
and demonstrate a real interest in applied statistics, experience in SAS is
required.
This could either be a 6 or 12 months contract.
Working in a very nice city with good connections to the UK.
If you are interested please get in touch for further details.
Kind regards
Mick
Mick Messing
Biometrics Recruitment
Key People Ltd
Adelaide House, Adelaide Street,
St. Albans, Herts AL3 5BE
Tel: 01727 817 609 (Direct Number)
Tel: 01727 811 634 (Switchboard)
Fax: 01727 856 594
Email: [log in to unmask]
Website: http://www.keypeople.co.uk
------------------------------
Date: Wed, 20 Oct 2004 13:01:10 +0000
From: Cathy Smith <[log in to unmask]>
Subject: relative risk or odds ratio
Dear All,
If I have an option of using RR or OR, which one should I use?, Which one is
more conservative?.
Cathy
_________________________________________________________________
Get ready for school! Find articles, homework help and more in the Back to
School Guide! http://special.msn.com/network/04backtoschool.armx
------------------------------
Date: Wed, 20 Oct 2004 15:53:31 +0100
From: "Cutajar Bezzina, Audrienne" <[log in to unmask]>
Subject: Short Course in "R"
Apologies for cross-postings
=20
There are just few places remaining for the course "R". R is
increasingly becoming more popular in statistics, and it can be
downloaded for free at http://cran.r-project.org/ =20
=20
The course runs from the 4th November 2004 to the 5th November 2004.
Lunch will be provided, and there are also some bursaries which would
enable the participant to attend the course free of charge. Travel
would also be reimbursed. =20
=20
For more information and a registration form, see:
=20
http://www.cas.lancs.ac.uk/short_courses/
=20
If you require further assistance, please contact
=20
Angela Mercer,
Centre for Applied Statistics,
Fylde College,
Lancaster University,
Lancaster,
LA1 4YF, U.K.
Tel: +44 (0)1524 593064.
Fax: +44 (0)1524 593429.
E-mail: [log in to unmask] <mailto:[log in to unmask]>=20
=20
------------------------------
Date: Wed, 20 Oct 2004 13:10:43 -0400
From: Regina Malina <[log in to unmask]>
Subject: Question on Clustering
Hi everyone,
I have a question on variable selection when clustering.
Should I remove highly-correlated variables when I perform clustering
analysis? Does it make a difference if I do not? Is there a rule of thumb on
how high the correlation should be in order to remove the variable?
Thank you in advance. Regina
------------------------------
Date: Wed, 20 Oct 2004 17:46:09 +0100
From: Claus Mayer <[log in to unmask]>
Subject: REMINDER: RSS HIGHLANDS GROUP MEETING, 21/10/2004: John Norrie
Dear all!
The local RSS Highlands group will have it's first talk of the new season on Thursday, 21st of October (Details are given below). Please pass this information on to anybody who might be interested.
Please also note that the webpage of the Highlands group has moved to:
http://www.bioss.sari.ac.uk/RSSH/
Regards,
Claus-D. Mayer (local group secretary)
-------------------------------------------------------------
TITLE: "Randomised controlled trials (RCT) post EU Clinical Trials Directive - some issues for statisticians"
SPEAKER: John Norrie, Director, Centre for Healthcare Randomised Trials (CHaRT), Health Services Research Unit, Aberdeen University
VENUE: Room WB27,Health Services Research Unit, Polwarth Building,
Foresterhill
TIME: Thursday 21st October at 5pm (tea and coffee from 4.45pm)
ABSTRACT:
The EU Clinical Trials Directive, along with other legislation and guidance, has given an exacting regulatory, ethical, and legal environment in which RCT are now conducted. This talk will give some observations on statisticians roles, from the perspective of an experienced triallist and statistician who has been involved in many RCT, both large and small, both publicy funded and ommercial, before and after the implementation of EU/CTD. The talk will include discussions of risk assessment of trials, some aspects of regulatory advice on both the design and analysis of RCT (via the Points to Consider series), and some thoughts on statisticians involvement in the monitoring of RCT.
--
***********************************************************************************
Claus-D. Mayer | http://www.bioss.ac.uk
Biomathematics & Statistics Scotland | email: [log in to unmask]
Rowett Research Institute | Telephone: +44 (0) 1224 716652
Aberdeen AB21 9SB, Scotland, UK. | Fax: +44 (0) 1224 716608
***********************************************************************************
------------------------------
Date: Wed, 20 Oct 2004 14:14:14 -0400
From: Art Kendall <[log in to unmask]>
Subject: Re: Question on Clustering
A great deal depends on the substantive nature of your data and the
reason you are clustering. Usually the variables are considered fairly
independent. In many contexts, it is common to use some form of factor
scores. I know of no rule of thumb for what is "highly correlated".
Clustering has a great deal of art to it. You would want to try several
approaches to see if the results are very different. You might also use
something like discriminant function analysis, remembering that tests as
such lose most meaning when you use the same variables as those in the
clustering. Many of the parts of the listing from a package like SPSS
can help give you insight into the different results.
Art
[log in to unmask]
Social Research Consultants
University Park, MD USA
(301) 864-5570
Regina Malina wrote:
>Hi everyone,
>I have a question on variable selection when clustering.
>Should I remove highly-correlated variables when I perform clustering
>analysis? Does it make a difference if I do not? Is there a rule of thumb on
>how high the correlation should be in order to remove the variable?
>Thank you in advance. Regina
>
>
>
>
------------------------------
Date: Wed, 20 Oct 2004 15:05:06 -0400
From: Isaac Dialsingh <[log in to unmask]>
Subject: Re: Medical stat
At the end of a contingency table in a medical paper, I am seeing two =
Chi-square test statistics over 2000 vs 2001 for each category of the =
column variable. Can anyone enlighten me as to what test is taking =
place?
Isaac Dialsingh
------------------------------
End of allstat Digest - 19 Oct 2004 to 20 Oct 2004 (#2004-245)
**************************************************************
|
