• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

ANALYTICAL SERVICES

OBJECTIVES

The Analytical Services program provides methodological expertise in statistical analysis to underpin the production of high quality statistics, to manage and mitigate statistical risks, and to exploit opportunities presented by the emerging information economy. Activities that support statistical production and the ABS transformation program are key focus areas for the medium-term. These activities include supporting analysis of time series data, data integration and confidentialisation methods, and innovative analytical methods for emerging data sources (e.g. big data).

The program contributes to the delivery and continued improvement of ABS statistical outputs by providing analytical products and methodological advice to both users and producers of statistics.

OUTPUTS

The main outputs of the program are:

• ongoing seasonal adjustment of critical time series used by Australian Government economic policy agencies, state and territory governments, and local government service delivery agencies
• methods, tools and practices that improve the quality or reduce the cost and time taken of current statistical production, and advance new directions in data access, integration and analysis, with particular emphasis on emerging data sources (administrative, transactional, and sensor data)
• methodological advice and assistance to users and producers of statistics within and external to the ABS in areas such as data linking, disclosure control, longitudinal data analysis, small area estimation, time series analysis, demographic analysis, and econometric modelling.

• conduct a big data feasibility study in harnessing satellite imagery data for agricultural statistics - due December 2015
• develop a prototype next generation business register that can serve as a flexible and scalable data integration spine for heterogeneous economic data - Phase 3 due March 2016
• evaluate new methods (e.g. model-based methods) and tools for seasonal adjustment to contribute to the ABS's transformation program - due June 2016
• develop methodology for the integration and confidentialisation of multiple datasets - due December 2016
• develop and implement new approaches for the confidentialisation of business microdata - due December 2016
• develop methodology for dynamic and longitudinal microsimulation in demographic analysis - due June 2017
• support collaborations on harnessing big data, in particular, with:
  • Statistics New Zealand (SNZ) on using semantic web methodology - ABS will help SNZ create a flexible analytical platform for the NZ Linked Employer Employee Dataset (LEED) (which provides statistics on jobs, worker flows, earnings, etc.) and SNZ will advise the ABS on how to productionise an Australian LEED and assess the potential analytical use - due June 2016
  • task teams under the UN Global Working Group on Big Data for Official Statistics, working to make better use of satellite imagery and geospatial data for official statistics by reporting on solutions to methodological, IT and privacy challenges and undertaking pilot studies on the use of this data - due December 2017
  • the Australian Research Council Centre of Excellence for Mathematical and Statistical Frontiers, to deliver research and outcomes which address the business problems (for areas such as big data, data integration and small area estimation) of industry partners including the ABS - until 2021
• develop methodology for longitudinal analysis, visualisation methods and tools for exploratory data analysis - due June 2017
• conduct semantic web demonstration projects using the graphically linked information discovery environment to improve the ABS's analytical capability in representing, transforming, integrating and analysing complex, multidimensional data from diverse sources - ongoing.