Menu
Log in


Events calendar

Upcoming events

    • 27 Sep 2021
    • 28 Sep 2021
    • Strasbourg, France

    Mathematics without Borders

    The Centennial of the International Mathematical Union         

    celebrates the centennial of this historic event.  Up-to-date information regarding the conference can be found here




    • 28 Sep 2021
    • 4:30 PM - 6:00 PM
    • Zoom
    Register

    Please join us online for the September Queensland Branch Meeting! There will be a branch meeting at 16:30 followed by a seminar-discussion at 17:00.

    When: 4:30 PM - 6:00 PM (AEST)

    Tuesday 28th September 2021

    Location: Online - Zoom

    Statisticians and the reproducibility crisis

    Abstract: Statisticians play a key role in almost all scientific research, and so are also key to tackling the reproducibility crisis which is undermining the value of much scientific research. Our role may be particularly important given that many problems with irreproducible papers are due to problems in the study design, statistical analysis and interpretation of results. We must promote more efficient, replicable and credible science, starting today, by being the role-model statisticians that we need for tomorrow.

    Our four speakers will give their perspective on statistics and the reproducibility crisis, and also give their opinions on how statisticians can help improve the quality of research. We hope to have at least 20 minutes for discussion after the speakers' presentations, so please come along and share your thoughts on this hugely important topic.

    Speakers:

    Dr Sabine Hoffmann is a postdoctoral researcher at the Institute for Medical Information Processing, Biometry, and Epidemiology, Ludwig-Maximillans-University of Munich. Sabine’s research projects investigates the effects of fishing for significance on statistical inference, and methods that allow handling of results variability when applying alternative analytical strategies. Sabine conveys these ideas in her consulting activities in order to raise awareness on how common research practices can have devastating effects on the replicability of research findings.

    Dr John Maindonald is a retired visiting research Fellow at the Centre for Mathematics and Its Applications, Australian National University (ANU). John has 50 years of experience collaborating and consulting with organisations and scientists across a wide range of application areas. Some of his research interests include statistical computation, statistical perspectives on data mining, use of R for practical data analysis, research planning and population genetics.

    Dr Teresa (Terry) Neeman is a biostatistician with the Biological Data Science Institute, ANU College of Science. She has more than two decades of experience as a biostatistician working with biologists, biomedical researchers and clinicians in both industry and academia. Terry enjoys the statistical challenges associated with all aspects of experimental work; from experimental design to data exploration and statistical modelling. Over the last several years, she has worked more closely with bioinformaticians, and has developed a strong interest in tools for visualisation and inference of highly complex data.

    Prof Adrian Barnett is the Program Lead – Statistics in Health Services at the Australian Centre for Health Services Innovations, QUT. He has spent more than 21 years working as a statistician, specialising in biostatistics, health services research and the research funding. He is particularly passionate about the application of statistics to health and the use of data analysis to develop cost-effective interventions within the health sector.

    Moderator: A/Prof. Dimitrios Vagenas


    • 28 Sep 2021
    • 5:30 PM - 6:30 PM (AEST)
    • Virtual via Microsoft Teams

    SSA Canberra invites you to its September branch meeting, which will be joint with Canberra Data Scientists

    Times (+/- some standard errors):  

    5:30pm – 6:30pm AEST (please note the earlier starting time at request of the speaker): Presentation on MS Teams

    RSVPNo RSVP will be required for this meeting, and please see below for the link. If you are worried MS Teams will encounter teething problems on your machine, please try out the meeting link at the bottom of this email, and if you have any questions that the internet cannot resolve (please also see this help page), please email Francis Hui at ssacanberra@gmail.com.

    SpeakerMs. Xiaoyan LuActing Executive Director, Australian National Audit Office (ANAO)

    Title:  Using Data Analytics for Audits

    AbstractIn this talk, Ms. Lu will share 

    • how data analytics is applied in the audit context;
    • her experience and lessons learned of establishing data analytics teams and delivering data analytics strategy in government agencies; and
    • what capabilities that employers are after when recruiting data scientists and analysts.  

     

    BiographyMs. Xiaoyan Lu has over 20 years’ experience in research and analytic industry and is an Acting Executive Director of the System Assurance and Data Analytics (SADA) Group at the Australian National Audit Office (ANAO). SADA provides IT audit and data analytics support and advice to all audit service groups within the ANAO. She joined the ANAO in April 2018 to build the Data Analytics team and deliver the Data Analytics Strategy 2018-20. She was awarded the Australia Day Achievement Medallion in 2020 in recognition of her leadership in transforming the delivery of audit outcomes through data analytics, commitment to development and mentoring of staff and her dedication and commitment to innovation and excellence. Prior to that, she was a Director at the Australian Sports Commission leading initiatives on data driven decision making in the sport sector, and worked there for 11 years. Before joining the public sector, she had worked in the academic, NGO and private sectors for providing research consultancy for eight years. Xiaoyan received a Bachelor degree in Science from Beijing Normal University, a Master degree in Environmental Science from Peking University, and a Master degree in Applied Statistics from Australian National University.


    **Microsoft Teams meeting link**

    Join on your computer or mobile app

    Click here to join the meeting

    This will bring up Teams in your default browser, even if you do not have the Microsoft Teams software installed.

    Note that, unlike Zoom, everyone by default sits in a waiting room and only comes in when the organiser comes in. 

    If you encounter any issues with MS teams and cannot find a solution online (please also see this help page), you can email Francis at ssacanberra@gmail.com or Yanchang Zhao at Yanchang.Zhao@data61.csiro.au

     

    Or call in (audio only)

    +61 2 9053 9802,,96171836#   Australia, Sydney

    Phone Conference ID: 961 718 36#

     

    For more information on joining Microsoft Teams meetings

    Learn more

     


    Website links

    https://statsoc.org.au/Canberra-Branch-meetings


    • 29 Sep 2021
    • 6:00 PM - 7:00 PM (AEST)
    This month we are very pleased to have Prof. Hanlin Shang from Macquarie University to present his recent work about sieve bootstrap and functional time series.
     

    Please note for security reasons, you will need to register in advance for this meeting: https://macquarie.zoom.us/meeting/register/tZEvceuhrT0tE9PY0vtRKVWW8LNbX0hMmBsg

    After registering, you will receive a confirmation email containing information about joining the meeting.

    Any questions, please feel free to contact: secretary.nswbranch@statsoc.org.au

    Date: Wednesday, 29th September 2021

    Time: 6:00pm - 7:00pm (AEST)

    Prof. Hanlin Shang
    Macquarie University, Sydney

    Sieve bootstrap memory parameter in long-range dependent stationary functional time series

    Abstract:
    We apply a sieve bootstrap procedure to quantify estimation uncertainty of long-memory parameter in stationary functional time series. To estimate the long-memory parameter, we use a semiparametric local Whittle estimator, where discrete Fourier transform and periodogram are constructed from the first set of principal component scores, via a functional principal component analysis. The sieve bootstrap procedure uses a general vector autoregressive representation of the estimated principal component scores. It generates bootstrap replicates that adequately mimic the dependence structure of the underlying stationary process. For each bootstrap replicate, we first compute the estimated first set of principal component scores and then apply the semiparametric local Whittle estimator to estimate the memory parameter. By taking quantiles of the estimated memory parameters from these bootstrap replicates, we can construct confidence intervals of the long-memory parameter. As measured by coverage probability differences between the empirical and nominal coverage probabilities at three levels of significance, we demonstrate the advantage of using the sieve bootstrap in comparison to the asymptotic confidence intervals based on normality.

    Biography:
    Hanlin is a Professor of Business Analytics at the Department of Actuarial Studies and Business Analytics, Macquarie University. Hanlin’s research interests include inference, modelling and forecasting functional time series.
    • 12 Oct 2021
    • 6:00 PM - 8:00 PM
    Register

    Statistical Society of Australia Victorian Branch is honoured to have Professor Andrew Forbes as our Belz lecturer for 2021 to talk to us about:

    Tales of clinical trials, tribulations, twists and turns

    In this era of COVID-19 we constantly hear about case numbers of people infected, vaccines and epidemic modelling, and, particularly early in the epidemic, randomised clinical trials of treatments for COVID-19-infected patients. In my experience, the general public’s perception of clinical trials is that they are limited to evaluation of new drugs or treatments for particular conditions. Arguably, among some statisticians (including an earlier version of myself), there may be a perception that the statistical issues and analyses of randomised trials are formulaic and rather routine, due to the reliance on the virtues of randomisation. However, this is certainly not the case and there is a fascinating world of randomised trials both inside and outside of drug development. In this talk I’ll focus primarily on trials outside of drug development and discuss my experience with a diverse array of such trials and some unanticipated twists and turns that needed addressing from a statistical or practical perspective.

    Speaker biography


    Professor Andrew Forbes is the Head of the Biostatistics Unit in the School of Public Health and Preventive Medicine at Monash University. He completed his undergraduate degree in mathematical statistics at Monash University, then went on to complete a PhD at Cornell University, USA. After a short stint as a postdoctoral research fellow in the pharmaceutical industry, he returned to Australia and joined Monash University where he has been since. His primary activities involve methodological research, particularly in the area of cluster randomised clinical trials, collaborating with researchers on clinical and public health research projects, including a variety of randomised trials, and teaching in the Master of Biostatistics program. He was recently elected as a Fellow of the Australian Academy of Health and Medical Sciences for his contributions to biostatistics and health research.

    Social event

    A social event will follow soon after the lecture with details to come soon. 

    • 25 Oct 2021
    • 1:00 PM - 2:00 PM (AEDT)
    • Online
    Register

    SSA peer-review seminar and panel discussion

    Are you curious about the peer-review process used by academic journals? Have you been asked to review a paper for a journal but not sure where to start? Do you want to improve your own academic writing and chance of getting published?

    In this seminar, Dr Myra McGuinness will provide a general overview of the peer-review process. This will be followed by a panel discussion on reviewing statistical methodology papers with Prof Kate Lee, Prof Andrew Forbes and Prof. Alan Welsh. This session is targeted at students and early-career statisticians, but we welcome anyone with questions about peer review to join.

     

    Dr Myra McGuinness: