Seasonal adjustment: 2015 update

This regular article reports on the annual review carried out in 2014 of the seasonal adjustment of the Bank of England’s money and credit data and other series. It also provides an update on other seasonal adjustment workstreams.

Published on 02 June 2015

Bankstats article

By Ross Meader

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 2014 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 selected series on a regular basis. 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;
- effects of 2002 Golden Jubilee and 2012 Diamond Jubilee; and
residual seasonality;

direct versus indirect adjustment (selected series only). Some series can be adjusted either directly or indirectly (as the sum, or the difference, of their seasonally adjusted components). The method chosen depends on various properties of the series and its components, and the relationships between these. The Bank continues to monitor the adjustment structures of key series to ensure that these remain appropriate.

Results of 2014 review

In total, 111 published series were reviewed in 2014, as detailed in Table 2. This resulted in changes in the seasonal adjustment settings for 40 series, as detailed in Table 3.

33 seasonal series were published for the first time, as follows:

Bankstats Table A5.4 now includes 30 new series, published on a seasonally adjusted basis. These series cover approvals and cancellations by monetary financial institutions and other specialist lenders. All but three were judged to be seasonal. These were first published in the December 2014 data.
following the separate publication of consumer overdrafts and loans excluding overdrafts, three series were reviewed for the first time. These series are LPMZ5G6, LPMZ5G7 and LPMZ5GI. These were first published in the February 2015 data.

Some series were reviewed for specific issues, as follows:

loans securitised to UK-resident SPVs switched from lending by ‘Other Specialist Lenders’ to lending by MFIs in January 2010.5 This caused a break in affected series and a change in the pattern of seasonal adjustment. The method used to seasonally adjust three of these series was changed to be adjusted separately before and after January 2010. These series are LPMB3SE, LPMVWAY and LPMB4TE.
the total public sector contribution to M4 lending (published in Bankstats Table A3.2) was changed from seasonally adjusting flows series to deriving flows from seasonally adjusting amounts outstanding series. This series is LPMVQRO. This was implemented in the April 2014 data.

Ongoing work

In addition to the regular review process, the Bank also conducts more specific work on issues relating to seasonal adjustment:

following the switch to X-13ARIMA-SEATS in June 2014, the Bank reviewed an alternative weekday/weekend regressor. The weekday/weekend regressor is referred to as the ‘td1coef’ regressor in X-13ARIMA-SEATS. Where series have a trading day effect, this regressor has been chosen for a majority of series, as it produces a preferred seasonal adjustment compared to the conventional six-coefficient trading day regressor.
changes to the treatment of loan transfers and lending to housing associations from this month may have some small effects on the seasonal adjustment of relevant series. The Bank aims to review these series over the next year, as part of the 2015 annual review process.

To view all related charts and Annexes, please download the full article:

Download PDF