STATS MATTERS & EVENTS

17 June 2021

Dear {Contact_First_Name}, 

Have you noticed that over the last few weeks, when you run into friends, family or colleagues, the question inevitably comes up: “Have you had your shot yet?”

The conversation then usually goes to whether or not you are one shot or two shots into being vaccinated for COVID-19, and then of course to the most important question: “Which vaccine did you get?”, with the Pfizer-BioNTech recipients sometimes coming across a bit smug when they reveal their vaccine brand to us AstraZeneca recipients (ok – now everyone knows I’m 50+ years old).

Like many of you I am still waiting for my second short, so I found it very interesting when I came across this article: “Mixing vaccines may boost immune responses”, written by Gretchen Vogel and published last week on the website for the American Association for the Advancement of Science.

Gretchen explains in her article that three recent studies showed that strong immune responses were produced when one dose of the vaccine made by AstraZeneca was followed up with one dose of the Pfizer vaccine. Two of the studies even suggest the mixed vaccine response will be at least as protective as two doses of the Pfizer- product. You can read the full article here

I’m not sure this article will be enough to convince your GP when you go for your second dose, but the outcome of this research should relieve any panic at the thought of providers running out of “your” brand of vaccine.

Marie-Louise Rankin
Executive Officer, SSA

Read newsletter in your browser

What Are the Odds? Even Experts Get Tripped Up by Probabilities

This article by Jo Craven McGinty, published on 11 June 2021 in the Wall Street Journal, explains how people tend to rely on anecdotal evidence when making decisions, rather than considering the numbers. The article goes on to highlight that this tendency can have unfortunate consequences for people’s finances and health. There has been evidence that people who are less numerate adopt fewer healthy behaviors; they are 40% more likely to have a chronic disease; they end up in the hospital or emergency room more often; and they take 20% more prescription drugs, but are less able to follow complex health regimens.

Read the full article here

Early Career Bayes Seminar Series*:
Bayesian hierarchical stacking—All models are wrong, but some are somewhere useful

Presented virtually and in person by Yuling Yao (Columbia) on

Wednesday, 23 June 23, 2021, 9:30 AM – 11:00 AM AEST

*Early Career: The organisers want to highlight great research from early career researchers as opportunities for ECRs to present their work to broad audiences has declined.

About this seminar: 

Stacking is a widely used model averaging technique. Like many other ensemble methods, stacking is more effective when model predictive performance is heterogeneous in inputs, in which case we can further improve the stacked mixture with a hierarchical model. In this talk Yuling will focus on the recent development of Bayesian hierarchical stacking: an approach that locally aggregates models. The weight is a function of data, partially-pooled, inferred using Bayesian inference, and can further incorporate other structured priors and complex data. The presenter will also discuss some theory bounds: when and why model averaging is useful; what model dissimilarity metric is relevant to Bayesian ensembles.

Many thanks to the sponsors:

QUT Centre for Data Science

ARC Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS)

Statistical Society of Australia (Queensland Branch)

Find out more and register here

The Centre for Data Science at QUT is hosting a Datathon!

The QUT CDS Decode Datathon is open to everyone (undergrads, postgrads, early career researchers, industry professionals etc.): whoever wants to experience and solve real-world data science problems and challenges, as well as learn new skills in a data science team setting. The Datathon will involve 4 days of on campus activities (or optional online participation for those in remote circumstances), as well as regular check-ins with mentors. The overview of the program is as follows:

  • ·5 July: Welcome, Industry Lightning Talks, Team Formation and Meet the Mentors
  • 6 July: Training in Data Science, Business Analytics and Twitter Analytics
  • 7 July to 23 July: Team Work and Check-ins
  • 26 July: Pitch Coaching, Data Visualisation and Data Science for Impact
  • 30 July: Pitch Practice, Pitch Showcase and Awards Session

All participants will receive Certificates of Completion, with outstanding teams winning the Decode Datathon trophies. There will also be opportunity to attend some optional social events with the AMSI Winter School that are happening at QUT during the 4 weeks.


For more information and to register your interest, please click here

Abel Prize 2021

The Norwegian Academy of Science and Letters hereby calls for nominations of candidates for the Abel Prize 2021.

The Abel Prize recognizes outstanding scientific work in the field of mathematics, including mathematical aspects of computer science, mathematical physics, probability, numerical analysis and scientific computing, statistics and applications of mathematics in the sciences.

The Abel Prize amounts to NOK 7,5 million.

Nominations are now open. Your nomination should be accompanied by a description of the work and impact of the nominee/nominees, together with names of distinguished specialists in the field of the nominee/nominees who can be contacted for an independent opinion. When nominating it is a requirement to take into account that the nominee has adhered to general guidelines for research ethics.

Your letter of nomination should be sent no later than 15 September 2021.


For further information and the nomination form please click here

Do you have recent statistics or data science postgraduate research to showcase and would like to win some prize money?

Then you’re in luck, as SSA Canberra are holding a final call for nominations for this year’s Dennis Trewin prize! If you have interesting research that is of interest to statisticians and/or data scientists, we strongly encourage you to apply. Eligible applicants that were unsuccessful in previous years are more than welcome to re-apply this year.

Eligibility criteria:

  • At least 12 months into a postgraduate research degree or graduated from a postgraduate research degree in the last 36 months (at the time of application).
  • Undertaking/undertook the research degree from a university with a campus within the ACT or regional areas of NSW outside Newcastle-Sydney-Wollongong.

Please send an expression of interest to ssacanberra@gmail.com by Wednesday 30th June, providing a brief summary of your research topic.

Shortlisted applicants will be invited to record a 15-20 minute talk on their research. The winner will receive a $1,000 cash prize and an invitation to present their research virtually as part of SSA Canberra's branch meeting in late October.

If you have any further questions regarding the nature/format/scope of the prize, please contact ssacanberra@gmail.com

Regards,

DT prize committee (SSA Canberra Council)

--

Statistical Society of Australia (SSA) Canberra Branch

Representing statisticians throughout the ACT and regional NSW

ssacanberra@gmail.com | https://statsoc.org.au/Canberra 

The Vic Branch of SSA is proud to present   

Julia for Statistics and Data Science -   Your first steps In statistics and integration with your current platform

8 Jul 2021– 9 Jul 2021 via Zoom

presented by Associate Professor Yoni Nazarathy from the School of Mathematics and Physics at The University of Queensland.

This two day workshop aims to enable R users and other data scientists to incrementally incorporate Julia in their workflow. After an introduction of Julia basics, the workshop focuses on the creation of a simple, yet computationally demanding simulation example. This Julia code is then incorporated in R and Python, illustrating how users may create new performant statistical software using Julia while maintaining existing code base in R or Python. With this exploration, participants will learn not just how to use Julia, but also how to integrate Julia into their day to day statistical analysis which may involve R or Python legacy code.

To learn more or register, please click here.

ECSSC2021 Video Competition 

Submissions will open shortly for ECSSC2021 Video Competition. Working on any research? Put a video together and demonstrate your ability to concisely disseminate your research to a wider audience. Any student or early career statistician (within 5 years of graduation) in a statistics related field is welcome to enter the competition (sorry – previous winners are excluded). There is no entry fee!

For inspiration, please consider viewing previous submissions and winners. All submitted videos have been posted in the SSA ECSSN youtube channel of Statistical Society of Australia.


Attention AStat Accredited Members

We are seeking expressions of interest from AStat accredited members wanting to join the Accreditation Committee. Get involved and help other members attain accreditation! Keep in mind that you would not be accredited yourself if it were not for the dedicated volunteers on SSA’s Accreditation Committee.

Committee tenure is usually for three years, but some committee members enjoy their time on the AC so much that they extend to six years! Meetings are held via Zoom every six weeks or so; sometimes just every other month. Accreditation applications are uploaded to Dropbox and the committee members are asked to read them prior to the meetings and then give their view on whether candidates meet SSA’s criteria for accreditation or not. In 2020 we received a total of 33 applications, many of them applications for reaccreditation which are usually very straightforward.

If you are interested, please contact me. If you would like to chat with someone on the committee before making a decision, please let me know and I will put you in touch with a committee member or the Chair.

I look forward to hearing from you.

Marie-Louise Rankin
SSA Executive Officer

SSA History Working Group: call for members

The SSA is calling for members interested in joining a working group to help record and preserve the history of the Society and of Australian statistics and statisticians. Initial work will involve compiling a set of profiles of prominent Australian statisticians, but additional activities focussed on recording the history of the SSA will be encouraged. Any members who are interested in joining this working group can get in touch with our Executive Officer.   

SSA Member Webinar: A Desingularized Mean Field Approximation

presented by Dr Susan Wei (DECRA fellow, University of Melbourne). 

18 Jun 2021, 3:00 PM – 4:00 PM AEST via Zoom

Unlike in regular statistical models, the posterior distribution over neural network weights is not asymptotically Gaussian. As established in singular learning theory, the posterior distribution over the parameters of a singular model is, asymptotically, a mixture of standard forms. Loosely, this means the parameter space can be partitioned such that in each local parameter set, the average log likelihood ratio can be made "normal crossing" via an algebraic-geometrical transform known as a resolution map. We leverage this under-appreciated result to propose a new mean-field variational family for Bayesian deep learning. Affine coupling layers are employed to learn the unknown resolution map, effectively rendering the proposed methodology a normalizing flow with the generalized gamma as the source distribution, rather than the multivariate Gaussian typically employed.

Susan Wei is a lecturer in the School of Mathematics and Statistics at the University of Melbourne. She currently holds a Discovery Early Career Researcher Award (DECRA) from the Australian Research Council (ARC). Her research interests include statistics, machine learning, and deep learning.

Due to a cancellation we now have availability again for the following workshop:

Introduction to Machine Learning for Health Data

8 Jul 2021, 9:00 AM – 9 Jul 2021, 5:00 PM, Flinders at Victoria Square,

Adelaide

If you are interested in detecting patterns associated with diseases and health conditions by studying large amounts of health records, but do not know where to start, SSA and Flinders University are proudly offering this 2 day course with Presenter Dr Oscar Perez-Concha, Centre for Big Data Research in Health, UNSW Sydney. This workshop introduces the basics for understanding and using machine learning algorithms.

Register here

Scholarships available for attending ANZSC2021, ECSSC2021 and AMSI Winter School

Several branches of SSA are offering financial support to fund  participation at the Australian and New Zealand Statistical Conference 2021, the Early Career and Student Statisticians Conference and AMSI Winter School. The Canberra Branch has extended the application deadline until 28 June 2021 and is now offering reimbursements to successful applicants who are already registered for the conference.

Schemes and deadlines vary slightly between branches. To find out what your branch is offering, please click on the appropriate blue box below.

QLD Branch
NSW Branch
VIC Branch
Canberra Branch

We'd like to thank  our other generous sponsors:

 
 
 

The ECSSC is offering several courses leading up to the conference:


Convex Optimization for Statistical and Machine Learning with CVXR

24 Jul 2021, 11:00 AM – 3:00 PM AEST via Zoom

Statistical Shape Analysis via Topological Data Analysis

25 Jul 2021, 11:00 AM - 3:00 PM AEST via Zoom

If you have news from the Australian statistical community to share in Stats Matters and Events, please get in touch with us! We love getting feedback too.


Statistical Society of Australia |  PO Box 213 Belconnen ACT 2616 Australia 

02 6251 3647 | www.statsoc.org.au