Statistical software interoperability

Thursday 14th November 2013

Statistical Software Interoperability

An afternoon meeting, preceded by the AGM featuring four presentations from leading figures in computational statistics giving talks on new developments that involve interoperability between statistical software packages

This is a joint meeting of the Statistical Computing Section (SCS) of the Royal Statistical Society (RSS) and the British and Irish region of the Biometric Society.

The meeting will be held at the RSS headquarters, Errol Street in London.

 Cost:

Member of Biometric/Royal Statistical Society: £20

Student/Retired  Member: £10

Non-Member: £35

 

Note that registration is now closed. If you wish to register late then please email Bill Browne (william.browne@bristol.ac.uk) to let him know and bring a cheque made payable to the Biometric Society with you to the meeting

Programme

 Document downloads for IBS members.
Join us now.
 
12.45 - 13.00Registration
 
13:00 - 13:30BIR AGM

Annual General Meeting of the BIR

 
13:30 - 14:15Statistical interoperability – what, why, when and how? The Stat-JR experience

Professor William Browne (University of Bristol)

In this talk I will try to give some motivation to what is meant by statistical interoperability and what motivated us to make it a part of our work on the Stat-JR package. I will describe how we moved from an ideal of a statistical model exchange to the system we finally developed and describe where such a system might be useful including examples of software comparison, parallel processing and the benefits of linking software together. To this end I will touch on the RunMLwiN and R2MLwiN functionality we have developed for Stata and R respectively. I will also talk about the difficulties in generalising interoperability between packages and describe from personal experience what can and cannot be easily done.

 
14:15 - 15:00The Interoperation of Statistical Software in a Commercial Setting

 Some Personal Experiences and Observations

Dr Ian Cox (JMP Marketing Manager, SAS)

SAS offers a very diverse set of software products and solutions. One of these is JMP, which aims to support 'statistical discovery' on the desktop. This presentation shares some of the motivation for, and challenges of, developing the interoperation of JMP with the rest of SAS. It then attempts to draw some wider conclusions about when such interoperation of statistical software is likely to be desirable and successful. 

 
15:00 - 15:30Coffee
 
15:30 - 15:40RSS Statistical Computing Section AGM
 
15:40 - 16:15GenStat and ASReml as components of the statistical toolkit

Dr Darren Murray (VSN International)

Developers of computing software (office suites, data bases etc) often seem to take the view that users should never need to go elsewhere to fulfill their needs, but can spend their entire computing lives within a single system. In statistics, it is harder to support that attitude (although some have tried!). Our subject, its methodology and applications, are so diverse and developing so quickly, that it would be arrogant and foolish for any one group to assume that they can cover the entire user-requirement. So statistical analysts may need to draw on different tools according to each situation. This talk will discuss some of the ways in which VSNi's systems, such as GenStat and ASReml, aim to support this way of working. Existing features range from the straightforward issue of file interchangeability through to the ability to act as an environment for other systems or to operate as statistical engines. There will also be a discussion of future challenges, including the ability to link systems to handle large data sets, and to provide secure analyses over the internet.

 
16:15 - 17:00There and Back again – A BUGS tale

 Mike K. Smith (Pfizer – Director of Pharmacometrics)

Using WinBUGS and OpenBUGS with other packages, such as SAS or R is relatively straightforward with the BUGS scripting languages. SAS macros or R packages can be used to set up and call BUGS scripts allowing interoperability between software. On the one hand this allows the user to prepare data for analysis in BUGS using familiar tools and to pass output from BUGS back into those systems for summaries and reporting but perhaps more importantly allows better reproducibility in analysis where the model, data preparation, analysis settings, summaries and diagnostics can be handled using a single script. This presentation will look at accessing BUGS from SAS and R using widely available tools.

 

Membership

Existing members can login below to view all site content. Lost password?

 
 

Other visitors might be interested to learn more about the benefits of membership.

Other events

11 Nov 20Estimating Abundance and Beyond
28 Oct 20Advances in statistical genomics
22 Sep 20Advances in Survival Analysis
02 Oct 19New perspectives on studying the effects of treatment on a time to event outcome
10 Jul 19 - 12 Jul 197th Channel Network Conference

© 2009-2024 Biometric Society, British and Irish Region | Admin | Read our cookie and privacy policy