Announcing the winners of the commendable Young Statistician Presentations

First place

Fu Shi Ching

Using Bayesian Networks to Identify Significant Crime Events in WA

Second Place

Duffin Connor

A Statistical Finite Element Method for Nonlinear PDEs

Third Place

Malone George

The L2 Method for Robust Estimation of Mixtures: An Application in Diagnosis

of Equine PPID

Poster Presentation Winner

Ng Kenyon

Predicting Oats Cutting (Flowering) Time with Semiparametric Additive Models

Congratulations from all of us at SSA!

On 30^th September and 1^st October 2020, the Western Australia branch held its 11^th biennial Western Australian Young Statisticians Workshop. The workshop aimed at supporting Young Statisticians and students to further their professional career in statistics. The workshop was held online this year to safely engage and assist Young Statisticians in collaborating with local professionals amongst the pandemic disruptions. This free workshop was held over two half-days, with a morning session on the first day and an afternoon session on the second day. On the first day, we had four invited speakers present as well as three Young Statistician talks. The morning concluded with a breakout session. Between sessions, attendees could view a poster presentation. On the second day, we held an afternoon session where three invited speakers presented, and four Young Statisticians gave oral presentations. The afternoon ended off with a breakout session followed by the announcement of the Young Statistician prize winners.

Seven invited speakers from industry and academia presented during the workshop, namely; Nazim Khan, Ross Taplin, Alex Maund, Anna Hayes, Emi Tanaka, Brenton Clarke andNoel Cressie. We gratefully thank our invited speakers for volunteering their time to talk and participate in breakout sessions.

We were delighted to hear from seven worthy Young Statisticians who expertly presented their work, namely; Sofina Begum, Joseph Sigar, Audrey Yeo, Alexander Rohl, George Malone, Connor Duffin, Shih Ching Fu and Kenyon Ng. The SSA awarded prizes to four excellent Young Statistician presentations. We also had a draw for survey respondents, with Parsa Amid being the lucky winner.

We would like to thank our sponsors, the Australian Bureau of Statistics, the Statistical Society of Australia and Murdoch Guild for supporting our Young Statisticians Workshop. Their generous sponsorship allowed us to have a free event with prizes and gifts for invited speakers.

A huge thanks goes out to our Young Statisticians sub-committee: Barbara Kachigunda, Alun Pope, David Urginov, Rick Tankard, Torben Kimhofer, and led by Deneegan Subramanian. We also thank our Young Statistician presentations judges Brenton Clarke, Ross Bowden and Berwin Turlach.

by Deneegan Subramanian

SSA WA Branch Young Statisticians Representative

Laura Edney, Research Fellow in Health Economics at Flinders University, spoke to our October meeting on her own work about the motivation for the use of Instrumental variables (IVs), the assumptions they make and how these can be appropriately tested either directly or through sensitivity analyses by examining the impact of assumption violation. This is the topic of her current research, following on from many years of work on various fields on health economics.

At the beginning of her talk, Laura described the IVs and their application to make causal conclusions from observational data when randomized controlled trials are not feasible. IV methods have been widely used to estimate the impact that spending on health has on health outcomes due, in part, to historic health outcomes representing an important unmeasured confounder that, if unaccounted for, will result in biased coefficients on spending in standard regression models. She addressed few examples of IVs and illustrated to determine variation that is exogenous in treatment and to estimate causal inferences. In particular, smoking and health outcome relationship, an IV cigarette price is not logically directly related to health. The only logical relation is an indirect one: price affects cigarette use that, in turn, affects health.

To illustrate the key concepts of IVs, Laura reviewed the application of IVs to estimating the impact of health spending on health outcomes focusing on eight publications that have estimated this relationship nationally. Using two-stage least squares estimation (2SLS), health spending has a significant impact on health outcomes in presence of IVs. IV quantile regression, an extension of IV approach, was also discussed about estimating across the full distribution compared to unweighted IV 2SLS estimates of the mean effect.

In conclusion, Laura mentioned two key questions to consider: (i) Is the instrument meaningfully related to the predictor variable? (ii) Does the instrument directly or indirectly

influence the outcome variable? Merits and demerits of IVs were also highlighted at the end of the presentation. The meeting was held in virtual platform Zoom video communication with almost 25 attendees in the meeting.

By Shahid Ullah

The Australian Mathematical Sciences Institute (AMSI) today announces the appointment of Professor Tim Marchant as its new Director, commencing in January 2021, following a comprehensive recruitment process.

Most recently Dean of Research at the University of Wollongong, Professor Marchant brings extensive academic and leadership experience to AMSI, and has a demonstrated passion for mathematical sciences having been President of the Australian Mathematical Society, Chair of Australia and New Zealand Industrial and Applied Mathematics (ANZIAM) and Director of the Mathematics in Industry Study Group.

“AMSI has a strong track record in working with schools, universities, industry, philanthropists, government and the community in shaping policy and skilling Australia for the future through its diverse programs across education, research and industry”, said Dr Adelle Howse, AMSI Chair. “Our Institute is delighted to welcome Professor Marchant as Director at a very exciting time for the mathematical sciences community, facing high demand from many sectors of Australia’s innovation ecosystem.”

“Professor Marchant is looking forward to working with AMSI membership, staff, stakeholders, government and funding partners in delivering AMSI’s mission: championing the mathematical sciences for Australia’s advancement.” added Dr Howse. “Tim shares AMSI’s passion to promote and develop mathematical sciences in education, research and industry.”

Completing his PhD at the University of Adelaide before starting an academic career at the University of Wollongong, Professor Marchant is an Applied Mathematician whose research stature in the fields of nonlinear waves, optics and industrial modelling is evidenced by 80 journal publications and supervision of 17 graduate research students to completion.

Congratulations to Basim Alsaedi from University of Tabuk, Saudi Arabia, and formerly from the University of New England where he completed his PhD, who has won the 2020 Dennis Trewin prize. Basim has received $1000 and a complimentary one-year SSA student membership for his efforts. 

The Dennis Trewin Prize, named after the former Australian Statistician, is awarded to a student from the ACT or regional NSW for their postgraduate research in statistics or data science. This year, five applications were received, from which shortlisted applicants were asked to submit a 15-20 minute recorded talk on their topic. Based on these submissions, we were privileged to have an external selection panel comprising of Dennis Trewin, Geoff Lee, Marijke Welvaert and Rachael Quill to decide on the recipient of the prize.   

As the winner of the Dennis Trewin prize, Basim presented his research on the topic of “Bayesian Modelling of Ion-selective Electrode Sensor Arrays” at the October Canberra branch meeting, which had an attendance of 29. Basim’s talk centred around the concept of the Limit of Detection (LoD) of a sensor – the lowest amount of a substance that can be reliably distinguished from absence of that substance for a given confidence limit. He proposed a Bayesian approach to determining the LoD that accounts for estimation uncertainty. This was then extended to the case where multiple sensors were being simultaneously used for substance detection.  

Following the talk, there was engaging discussion regarding the methodology used and the relevance of Basim's work to COVID testing. We thank Basim for presenting his fascinating research.  

Francis Hui

The pollsters did not perform well in the lead up to the 2019 May Federal election. All predicted a win by the Labour party (ALP) with margins from 2% to 4% of the two party preferred vote. In the event, the Liberal National Party (LNP) coalition won by 3.0%. Election and political polling are consequential and influential and play an important role in the democratic process. In recent times, at least one Prime Minister lost his job as a result of poor polling. Furthermore, policy decisions are influenced by opinion polls.

Both the Statistical Society and the Australian Market and Social Research Organisations (AMSRO) proposed reviews of the performance to determine why all the published polls incorrectly called the outcome at 2019’s Federal election and how methods can be improved in the future.

It was agreed that AMSRO would take the main responsibility for the review but that the Statistical Society would be closely involved.

Modalities

A panel was established under the leadership of Darren Pennay. Darren was the Founder and past CEO of the Social Research Centre. the other panel members were a mixture of political scientists, statisticians and polling experts. Several were Statistical Society members. They were assisted by an Advisory Board. John Henstridge was the Statistical Society nominee for the Board. The Chair was Dennis Trewin who is also a Statistical Society member.

AMSRO invited pollsters, media organisations and others who commission election and political polling to contribute to the inquiry. While the Inquiry Panel was grateful for the cooperation it received from the pollsters, the lack of access to data sets (which have been routinely provided to similar reviews overseas) and to detailed descriptions of the survey methods and statistical techniques used affected their ability to conclusively identify all the specific factors that contributed to the relative inaccuracy of the 2019 polls.

The Inquiry sought cooperation from the four pollsters responsible for these polls: Essential Research, Ipsos, Roy Morgan Research and YouGov (responsible for the YouGov Galaxy and Newspoll polls). Roy Morgan Research refused outright to co-operate.

Despite the limited information, a quite substantial report was prepared. It was released on 11 November and it and further information on the AMSRO Polling Inquiry can be found at: https://www.amsro.com.au/amsro-polling-inquiry/.

Findings

The report said that the polls erred in their estimate of the vote in a manner that was statistically significant and erred in the same direction and at a similar level. It also found that the source of errors lie in the polls themselves and was not a result of a last-minute shift in preferences among voters.

The first preference votes were either underestimated (LNP) or overestimated (ALP) because of inadequately adjusted, unrepresentative samples. The report said that it was very likely the polls were skewed towards over-representing more politically engaged and better educated voters and this bias was not corrected and as a result, the polls over-represented Labor voters.

The Inquiry Panel could not rule out the possibility that the uncommon convergence of the polls in 2019 was due to ‘herding’ although it may be due, in part, to their methodologies leading to similar weaknesses in the representativeness of their samples.  The absence of access to the data or the detailed methods means that this this remains a conjecture.

The Inquiry Panel also found that the reporting of the polls failed to meet the basic disclosure guidelines for editors and journalists set out by the Australian Press Council.

Recommendations

The Inquiry report has made 10 key recommendations aimed at improving polling methods and the disclosure thereof.  Recommendations include (1) calling for an enforceable Code of Conduct for Election Polling developed in consultation with experts including statistical experts; (2) improvements and greater transparency in polling methods, (3) better measures of uncertainty and (4) better education of the media and public on polling.

We believe there is an important role for statisticians and members of the SSA in particular in improving the quality of polls, both in the design of the surveys and in generating a productive debate about their performance. 

Postscript - 2020 US election polling

The accuracy of the US election polls has come under criticism.  Most of the criticism came early when it appeared that Biden might lose the election. As the count progressed with the inclusion of mail-in votes the difference between the poll predictions and the actual vote narrowed. In the end, how did the polls perform?

The following analysis is based on data from Real Clear Politics, one of the main poll aggregators in the US.

Our [Real Clear Politics] conclusions are:

• The performance of the polls was not great but better than generally perceived.
• At the national level, the poll average was a 7.2% margin to Biden compared with an expected final outcome of about 4.5%. At least 4 of the polls predicted the result within the margin of error.
• There was only one State where there was a polling miss – FLORIDA – where the average of the polls suggested a small win to Biden rather than a comfortable win to Trump.
• The average of the polls was very good in many of the Swing States and predicted the correct outcome in PENNSLYVANIA, MICHIGAN, MINNESOTA, NORTH CAROLINA, ARIZONA, and NEVADA, mostly quite close to the actual margin.  We are treating GEORGIA as an effective dead heat, but the average of the polls suggested a small win to Trump.
• The average of the polls was outside the margin of error in several of the rust belt States – OHIO, WISCONSIN, IOWA.

Like Australia and in the 2016 US elections, we expect the main reason for any lack of performance will be unrepresentative samples but that is conjecture at this stage. Polling is more difficult in the US because they have to predict election turnout which may vary between the candidates.

Dennis Trewin and John Henstridge

Science and Technology Media Release

Monday 9 November 2020

STA warmly welcomes Dr Cathy Foley as Chief Scientist

Science & Technology Australia has warmly welcomed today’s announcement that former STA President Dr Cathy Foley AO PSM FAA FTSE will take the reins as Australia’s next Chief Scientist.

Science & Technology Australia President Associate Professor Jeremy Brownlie congratulated Dr Foley on behalf of the STA membership representing Australia’s 80,000 scientists and technologists.

“Cathy is a superb choice for this role. She is hugely respected across the worlds of science and policy, a wise and clever leader, a generous mentor, and a skilled public communicator,” he said.

“All of those qualities will come to the fore in this moment of historic opportunity for science: to apply the centrality of science during COVID to so many other complex challenges.”

“We are also especially proud that she is a former President of STA, which is a crucial voice for the science community.”

Dr Foley was also the inaugural chair of the STA policy committee in its re-formation in 2017, deepening the sector’s policy engagement, and served as STA President from 2009-2011.

She is also a highly accomplished physicist and the current Chief Scientist at Australia’s national science agency CSIRO. Her research interests have focussed on solid state physics, quantum physics, and research translation.

The events of 2020 have highlighted the crucial role of bringing scientific advice to Government – an effort led by the Chief Scientist – and we know Cathy will excel in the role.

She was deeply involved in the STEM sector’s response to the bushfire crisis providing expert advice on the capabilities of industry and science and has had a key role in the Rapid Research Information Forum.

“STA also thanks outgoing Chief Scientist Alan Finkel for his work in creating the Science Policy Fellows program, setting up the RRIF and engaging widely.”

A LEGENDARY way to do observational data analysis at scale

The South Australian branch was pleased to have A/Prof Nicole Pratt to present at the august monthly virtual branch meeting.  A/Prof Pratt is an expert in biostatistics and pharmaco-epidemiology, specializing in the development of methodologies to study the effects of medicines and medical devices in linked health-care datasets.  She gave a talk on A LEGENDARY way to do observational data analysis at scale.

The talk was interesting and well-presented. A/Prof Pratt started with information on OHDSI (Observational Health Data Sciences and Informatics) coordinating center which works as a platform for international researchers to work collaboratively on a large-scale observational health dataset. In this distributive network, source data has transferred to standardized de-identified patient level data and you can perform analytics and produce results for publishing. She talked about ATLAS software and cohort Pathway package to perform collaborative studies and gave example of a hypertension study where data from 11 different databases in 4 countries (N= 250 million) was analyzed for different treatment pathways to treat hypertension. 

Later in her talk, A/Prof Pratt informed us that OHDSI collaborative launched a Large-Scale Evidence Generation and Evaluation across a Network of Databases (LEGEND) research initiative, aiming to generate evidence on the effects of medical interventions using observational healthcare databases and addressing the missing evidence from clinical trials. She defined ten principles of LEGEND, prescribing the generation and dissemination of evidence on many research questions at once, comparing all treatments for a disease for many outcomes, thus preventing publication bias and avoiding p-hacking. The Best-practice methods addressed measured confounding and control questions (questions where the answer is known) quantify potential residual bias. She gave an example of a study published in Lancet where they have looked at comprehensive comparative effectiveness and safety of first-line antihypertensive drug class, where every treatments contrast for every health outcomes (52 outcomes) with 9 different data sources. This study has enhanced the evidence around first line hypertension. She concluded her talk mentioning how OHDSI framework helping in generating evidence in a network of databases to assess consistency, by sharing open source analytics code to enhance transparency and reproducibility, but without sharing patient-level information, ensuring patient privacy. 

There were quite few questions from the audience after the talk, finishing the meeting at 7:10 pm.

By Aarti Gulyani

Australia’s peak body for science and technology has today welcomed four new members to its Board of Directors, representing strong diversity and breadth in STEM disciplines.

STA announced the following new appointees to its Board:

• Agricultural and food sciences – Mr Michael Walker, Soil Science Australia Executive Officer
• Biological sciences – Dr Tatiana Soares da Costa, Australian Society for Biochemistry and Molecular Biology (La Trobe University)
• Physical sciences - Professor Jodie Bradby, Australian Institute of Physics (ANU)
• Technological sciences - Dr Vipul Agarwal, Australasian Society for Biomaterials and Tissue Engineering, Royal Australian Chemical Institute (UNSW Sydney)

STA President Associate Professor Jeremy Brownlie warmly welcomed the new members of the Board, which oversees governance and sets strategy for the organisation.

“STA’s governance reflects our diverse membership, which collectively represents over 80,000 STEM professionals in Australia,” Associate Professor Brownlie said.

“Our incoming Board is genuinely diverse across the STEM disciplines represented, as well as across gender and culture.”

“Diversity of experiences and expertise is vital to strong Boards. It’s essential to robust decision-making, quality governance and visionary leadership for organisations.”

“On behalf of the organisation, I warmly welcome our incoming Board members, both those who have been newly appointed, and those who have been re-elected.”

Dr Tom Cresswell (Society of Environmental Toxicology and Chemistry Australasia), Jas Chambers (Australian Meteorological and Oceanographic Society), Marina Costelloe (Institute of Australian Geographers), Professor Adrian Barnett (Statistical Society of Australia) and Professor Rachel Burton (Australian Society of Plant Scientists) return to the STA Board for another 2 year term each.

STA’s new executive committee will be elected at the Annual General Meeting in late November.

For interviews, please contact STA Communications Manager, Zoya Patel on 0406 249 786.

More about our new Board members:

Dr Vipul Agarwal

Vipul is an NHMRC Fellow and Lecturer at the University of New South Wales (UNSW Sydney) who applies multidisciplinary research to develop bioimplants for spinal cord regeneration. He has strong experience working with industry and takes pride in advocating diversity and equity in science and scientific research.

Professor Jodie Bradby

Jodie is a physicist at the Australian National University with expertise in high pressure physics and the creation of new crystal phases of matter. Her group has made a type of silicon that can create a more efficient solar cell and a new form of diamond which is predicted to be even harder than regular diamond. Jodie is the Past-President of the Australian Institute of Physics and was a former AIP Women in Physics medallist. She has a strong interest in equity issues in STEM.

Dr Tatiana Soares da Costa

Tatiana is an early career biochemist and Group Leader at La Trobe University. She has won NHMRC Early Career and ARC DECRA fellowships to research antibiotics and herbicides to tackle the exponential rise in resistance. Tatiana was the founding President of La Trobe’s postdoctoral society. She is Editor of the Australian Biochemist magazine and the Australian Society for Biochemistry and Molecular Biology’s Chair of Communications and Science Advocacy.

Mr Michael Walker

Michael is Executive Officer of Social Science Australia and has spent his career in horticultural and agricultural professional associations. He has extensive experience in NFP membership, events management, board governance and project management. Michael is a former Chair of the Horticultural Training Council and of Primary Skills Victoria.   

Propensity score techniques in multiple treatments framework: the estimation of neighbourhood effect

The speaker for the September branch meeting was Margherita Silan, who is a postdoctoral research fellow at Padua University. Margherita was the winner of the Italian Statistical Society annual award for best PhD in Applied statistics. Her research interests include causal inference in multiple treatment frameworks, composite indicators and partially ordered set theory. Margherita’s research involved the estimation of the neighbourhood effect using propensity score techniques in a framework with many treatments, with an application to two health outcomes in Turin.

The neighbourhood effect is the independent causal effect of a neighbourhood on health and/or social outcomes. In order to compare the neighbourhood effect, it is important to make neighbourhoods comparable with respect to confounders. The Turin longitudinal study was used as the data source and outcomes considered were hospitalised fractures and incidence of depression or dementia. Neighbourhoods in Turin were represented as treatments, with at least 10 neighbourhoods included, depending on the partition methods.

Magherita compared two methods to estimate the neighbourhood effect; inverse probability of treatment weighting (IPTW) for multiple treatments and logistic regression with dummy indicators for neighbourhoods as covariates in the model. Simulations were performed to compare the methods with fracture as the outcome. Performance of the two methods depended on the scenarios simulated.

The neighbourhood effect was estimated for the outcome of dementia or depression using IPTW. An effect on mental health with respect to the mean was found, which was protective or harmful depending on the neighbourhood and gender. Based on her results, Magherita proposed a novel method: Matching on Poset based Average Rank for Multiple Treatments (MARMoT). The average rank summarises individual characteristics which are important for treatment allocation and is used as a tool to improve the balance of covariates between groups through matching. She described how to use the tool to achieve covariate balance and performed a simulation study to evaluate its performance. MARMoT improved covariate balance for 70 neighbouring zones in Turin and the average treatment effect in the treated changed considerably for many neighbourhoods.

The talk concluded with a Q & A session, including suggestions for future collaboration and requests to share her code.

By Lan Kelly