As the statistical agency of the Government of Australia, the Australian Bureau of Statistics provides key statistics on a wide range of economic, population, environmental and social issues, to assist and encourage informed decision making, research and discussion within governments and the community.
ACEMS and the ABS collaborate on cutting edge statistical methodological and applied challenges including remotely sensed data, data linkage, machine learning and recommender systems.
Collaborating with the experts within ACEMS has enabled the ABS to explore how statistical machine learning methods can influence the production of official statistics. Bringing together ABS staff and ACEMS academics in a collaborative environment ensures that the organisational culture and methodological knowledge are brought together to develop applicable and implementable insights, not only academic insights.”
Sybille McKeown, Australian Bureau of Statistics
ACEMS and the ABS have a long history of collaboration, which has been built on and broadened over the years. In 2019, we continued collaborating with the ABS on research, enjoyed a diversity of engagements including attendance at the ACEMS Annual Retreat, and marked the retirement of ACEMS Partner Investigator (PI) Dr Siu-Ming Tam from the ABS, wishing him farewell and welcoming new PI Dr Anders Holmberg.
The ABS has been an ACEMS Partner Organisation since inception and we count many successes delivered by this collaboration, including: contributions to the United Nations Global Working Group (UNGWG) on the use of big data and remotely sensed data for statistics, including innovative methodologies, reports and training workshops with attendees from around the world; PhD student co-supervisions; collaborative research workshops for identifying research challenges; technical training workshops for ABS staff; case studies, including on crop identification and crop yield; and the active involvement of ABS staff at the ACEMS Annual Retreat, affording an opportunity for ACEMS students and researchers to come together with the ABS, network, and discuss potential future collaborations and projects.
In 2019, the Centre’s collaboration with the ABS has focused on Statistical Machine Learning Methods research, which began in 2018 and is scheduled to be completed in 2020. This project explores the use of machine learning methods in the development of official statistics at the ABS. In this project ACEMS academics are working with the ABS Methodologies division to explore the application of Machine Learning Methods to areas such as: (1) the detection of cybersecurity attacks; (2) the prediction of occupancy on census night; (3) editing and correction using statistical machine learning; and (4) data collection protocols.
ABS representative Philip Bell attended the 2019 ACEMS retreat and presented within the Industry Panel at the main retreat.
Already, discussions are underway to explore new collaborations in 2020.
ABS’ Phillip Bell presenting to a diverse audience at the ACEMS Retreat.
In 2019, our long-term Partner Investigator for the Australian Bureau of Statistics (ABS), Dr Siu-Ming Tam retired after more than three decades of service at the ABS. Dr Tam has been a valued and engaged collaborator with ACEMS since the original ACEMS partnership agreement was signed in 2014. We thank him for his work in fostering collaborations with ACEMS over the past five years.
Dr Tam finished his full-time career with the ABS in September 2019 as a bureau stalwart and chief methodologist credited with many achievements in his three decades of service. Amongst these are his involvement in efforts to: publish ABS statistics online for free public use, which Dr Tam says “enables better decision making”; allowing use of ABS statistics under a creative commons licence to limit bureaucracy; and his collaboration with ACEMS. Read more about Dr Tam’s celebrated career with the ABS in this Canberra Times news article.
ACEMS continues to enjoy a productive working relationship with the ABS team and welcomed Anders Holmberg as the new PI for the ABS.
Dr Anders Holmberg joined the ABS in September 2019 as the Chief Methodologist. Dr Holmberg has a background as a survey methodologist, developing both theory and practice within data collection methodology, survey sampling theory and register based statistical methods.
Immediately prior to joining the ABS, Anders was the Chief Methodologist at Statistics Norway. He was also the Principal Statistician at Statistics New Zealand from 2014 to 2016, where he worked with traditional censuses and administrative data systems.
In 2012-13, Anders was responsible for the management of the Cambodian Socio-Economic Survey, as well as training staff throughout the statistical value chain. He enjoyed a 13-year career at Statistics Sweden (1999-2012), undertaking a variety of roles. Prior to this, he was a lecturer at Örebro University.
Dr Holmberg has a PhD in statistics from Uppsala University with the thesis "Essays on Model Assisted Survey Planning".