Bankstats article
By Katie Taylor
Introduction
Seasonal adjustment aims to identify, estimate and remove regular seasonal fluctuations and typical calendar effects (e.g. numbers of trading days in a month) from time series data. This article describes the 2016 annual review and summarises its results. It also provides an update on other seasonal adjustment workstreams.
Annual Review Process
The Bank of England reviews the seasonal adjustment of published series on a regular basis. The frequency of reviews of particular series will vary, based on the usage of the series. The annual review is divided into three phases spread across the year, as detailed in Table 1.
For each data series reviewed, the following issues are routinely considered:
- presence of seasonality;
- seasonal adjustment settings:
- choice of ARIMA model;
- calendar effects;
- outliers;
- seasonal and trend filters;
- residual seasonality; and
- direct versus indirect adjustment (selected series only)
Results of 2016 review
In total, 125 published series were reviewed in 2016, as detailed in Table 2. This resulted in changes to 36 existing published series (Table 3).
Of these, 32 series saw changes to their existing seasonal adjustment settings. In addition, 3 series not previously considered seasonal displayed seasonality when reviewed in 2016 and were assigned seasonal adjustment settings. There was also 1 previously seasonal series that was deemed to no longer be seasonal.
Furthermore, there are no newly published series that are being published on a seasonally adjusted basis as a result of this review.
Ongoing work
In addition to the regular review process, the Bank also conducts specific work on issues relating to seasonal adjustment. Two such issues are highlighted here:
The first is that the seasonal adjustment review has revisited the current seasonal treatment of household Cash ISA deposits. In July 2014 new rules were introduced, including an increase in thresholds and allowing transfers from Stocks and Shares ISAs to Cash ISAs. The impact in the data of these new rules was a significantly large inflow into cash ISAs in July 2014. With nearly three years of data we were able review the extent to which seasonality has changed.
The July 2014 increase had initially been attributed to a seasonal effect even though the effect had been in a different month to the usual seasonal effect (the usual seasonal month is April). This has been changed to treat the effect as irregular. This change in treatment has also corrected the seasonally adjusted level by around £10bn. These changes are due to take effect for April 2017 data, which will be published 31 May 2017, alongside this article.
Also, the Bank’s review of alternative trading day stock regressors continued. Until now, the bank has made use of user-defined trading day stock regressors but has been investigating whether the in-built trading day stock regressors, available in X-13ARIMA-SEATS, may be more appropriate. The results of this consideration suggested that there is no immediate benefit from altering these, however we expect to revisit this review in the future.
To view all related tables, please download the full article:
Download PDFFor questions relating to this article please contact dsd_ms@bankofengland.co.uk or call +44 (0) 20 7601 5361.