The ONS Time Series Analysis Branch (TSAB) provides expertise in time series analysis, in particular seasonal adjustment.
TSAB is responsible for the quality of all ONS seasonal adjustment and manages a rolling programme of annual seasonal adjustment reviews of statistical outputs across ONS.
TSAB also provides support and training for practitioners and users of seasonal adjustment across the Government Statistical Service and is the main point of contact for any questions or queries regarding seasonal adjustment and time series analysis in official statistics.
Areas of expertise: Seasonal adjustment, forecasting, benchmarking, interpolation, state space modelling.
Weather and Climate: The effect on official statistics
One of the areas that TSAB looks at is the effect of the weather and climate on official statistics.
For more details about why the guidance is needed and who it’s for, please see the Weather and Climate page.
Workshop for international time series experts: 6-7 March 2017
On 6th and 7th March 2017, the ONS Time Series Analysis Branch (TSAB) hosted a workshop for international time series experts, at the RSS headquarters in London.
The two day event was well attended by experts in the fields of official statistics and academia. Attendees travelled from National Statistics Institutes (NSIs) in Canada, France, Italy, United Kingdom and United States of America, the National Bank of Belgium, Eurostat and the Universities of Bristol and Lancaster.
Following the success of a similar meeting held at the RSS buildings in 2015 it was decided to put on a further meeting, this time focusing the content on the common time series issue Pre-Treatment.
This is loosely defined as adjustments for outliers and/or calendar effects prior to seasonal adjustment, but could include dealing with structural changes, selection of an appropriate spans of data etc.
Day 1 of the workshop started with each institute giving a short, high-level overview of time series issues they have encountered, or current research at their institute. Each institute was then invited to present in more detail any issues related to the pre-treatment of time series.
Day 2 kicked off with group discussions, based on four emerging themes from day 1: Outliers, Calendar Adjustments, Beyond Univariate and Robustness. Each group discussed a different aspect of pre-treatment and potential solutions to overcoming these in the future. The meeting was then wrapped up with some final presentations on high frequency data, which was noted as a potential theme for future meetings.
A detailed summary of each presentation, and associated discussion, can be found in the attached files.
The presentations from each institute can also be found here.
If you have any queries about any of the content please contact Bethan Russ.
Telephone – 01633 651737
Event presentations and minutes