Northern Ireland Quarterly Index of Production

Impact of Seasonal Adjustment Review
Theme: Economy
Geographical area: Northern Ireland
Date: June 2026

Introduction

This report summarises the impact of the periodic Seasonal Adjustment Review on the quarterly Northern Ireland Index of Production (IOP) estimates.

The quarterly Index of Production (IOP) provides a timely indicator of change in the output of the private sector production industries in Northern Ireland. This is defined as Standard Industrial Classification (SIC) 2007 sections B, C, D and E. Output estimates are calculated from the IOP aspect of the Quarterly Business Survey (QBS). The IOP has a sample size of approximately 1,200 businesses, covering all relevant companies employing 40 or more employees and those employing 0 to 39 employees and having a turnover of £10 million or more, along with a representative sample of smaller businesses. The sample frame for IOP is the Inter Departmental Business Register (IDBR), a register of all businesses registered for VAT and/or PAYE. More information on the IOP methodology can be found on the NISRA website.

A seasonal adjustment review for IOP was carried out in April 2026, by the IOP team within NISRA in conjunction with the Office for National Statistics (ONS), with the updated models implemented in the Quarter 1 2026 publication. The aim of the review was to ensure that the seasonal adjustment model utilised in each time series was appropriate and working well. The existing IOP seasonal adjustment models had been determined in a previous ONS review in June 2025.

Background

Economic output data can be affected by events throughout the year given that some business activity may be seasonal (for example, there may be greater demand for certain goods during or before the Christmas period). Output estimates from IOP are seasonally adjusted to account for such seasonal trends. Over time these seasonal patterns can change which necessitates periodic reviews of existing seasonal adjustment models.

Review Methodology

The 20 quarterly series reviewed are shown in Figure 1. The series name is a code that is used to refer to each series throughout this report, consistent with the names used in the previous review for the same series. The businesses which make up each series are identified by their UK Standard Industrial Classification (SIC) code which classifies business establishments and other statistical units by the type of economic activity in which they are engaged.

Figure 1 Series Reviewed April 2026
Series Name Description
IOP Overall Index of Production
ALL C Manufacturing sector (SIC07 Section C)
E Water supply, sewerage, and waste management (Inc. recycling) sector (SIC07 Section E)
ALL D Electricity, gas, steam and air conditioning supply sector (SIC07 Section D)
B Mining and quarrying sector (SIC07 Section B)
ENG Engineering and allied industries, Manufacturing subsector (SIC07 Divisions 26, 27, 28, 29 and 30)
TOTOTHER Total Other manufacturing, Manufacturing subsector (SIC07 Divisions 16, 17, 18, 19, 22, 23, 31, 32 and 33)
CA Food products, beverages and tobacco, Manufacturing subsector (SIC07 Divisions 10, 11 and 12)
CH Basic and fabricated metal products, Manufacturing subsector (SIC07 Divisions 24 and 25)
CECF Manufacturing of chemical and pharmaceutical products, Manufacturing subsector (SIC07 Divisions 20 and 21)
CB Textiles, leather and related products, Manufacturing subsector (SIC07 Divisions 13, 14 and 15)
Consumer Consumer goods are the final goods produced by industry which are intended for purchase by private consumers. These goods are consumed by market rather than used in the production of another good and are therefore closely linked to consumer demand and the factors which influence this
Intermediate Intermediate goods are those purchased by the industry as inputs into the final production of goods. This category would include materials (for example, cement, rubber, plastic, chemicals and electronics) which will ultimately be used to produce a good for consumption
Investment Investment goods (capital goods) are goods which enable production. For example, plant, equipment and inventories used to produce goods for consumption. Investment increases if business wish to expand or upgrade existing equipment
CC Wood & paper products & printing & reproduction of recorded media (SIC07 Divisions 16, 17 and 18)
CDCM Other manufacturing (SIC07 Divisions 19, 31, 32 and 33)
CG Rubber plastic & non-metallic mineral products (SIC07 Divisions 22 and 23)
CICJ Computer, electronic, electrical & optical products (SIC07 Divisions 26 and 27)
CK Machinery and equipment n.e.c (SIC07 Division 28)
CL Transport equipment (SIC07 Divisions 29 and 30)


Any exact additive relations that hold between series before seasonal adjustment are not guaranteed to be preserved between the seasonally adjusted series. Such relations, however, will still hold approximately.

From the date of the last review, there have been revisions to all data that were previously reviewed due to updates to the GVA estimates, (from 2019 to 2023), changes in the deflators, and an index rebase to 2023. Four quarters of additional data (Q1 to Q4 2025) have also been added.

Each series was reviewed using a seasonal adjustment program called X-13ARIMA-SEATS. The span of data used in the review was from Quarter 1 2005 to Quarter 4 2025. Each review included:

  • Assessment of whether the series is seasonal. Analysis of changes in economic output is complicated by regular effects associated with the time of the year and the arrangement of the calendar that obscure movements. For example, the demand for goods may increase during the winter or coming up to Christmas and this may obscure underlying movements in the overall trend. The purpose of seasonal adjustment is to remove variation associated with the time of the year and the arrangement of the calendar. This helps users to interpret movement in the series between consecutive time periods.
  • Choosing the appropriate decomposition type, that is, additive or multiplicative. In a multiplicative decomposition, the seasonal effects change proportionately with the trend. If the trend rises, the seasonal effects increase in magnitude, while if the trend moves downward the seasonal effects diminish. In an additive decomposition the seasonal effects remain broadly constant regardless of which direction the trend is moving in. In practice most economic time series exhibit a multiplicative relationship and hence the multiplicative decomposition often provides the best fit.
  • Calculating prior adjustments to be made to the series before seasonal adjustment. For example: temporary prior adjustments for outliers and level shifts; and permanent prior adjustments for trading days, Easter effects and seasonal breaks.
  • Selecting the ARIMA forecasting model. The purpose of ARIMA modelling is to identify systematic structural features in the history of the series. We assume that these features will continue to be present in the future and will use them to forecast future values. The ARIMA method provides a wide range of possible models, which have been found very effective in modelling typical socio-economic series showing trends, seasonality and business cycle effects.
  • Deciding the lengths of the seasonal and trend moving averages. Seasonal moving averages are weighted arithmetic averages applied to each quarter over all the years in the series i.e. a particular seasonal moving average is applied to each column of data. They are used by the X-13ARIMA-SEATS program to estimate the seasonal component of the series. The trend moving averages are weighted arithmetic averages of data along consecutive quarters. In general 9-, 13- or 23-term averages are used for monthly data and 5- or 7-term averages for quarterly data.
  • Reviewing the X-13ARIMA-SEATS diagnostics, both quantitative and visual. The quality of a statistical output should be determined by its performance against a range of attributes that together can be used to assess whether an output meets users’ quality criteria.


The first stage of a review is a “default” run where all the model and parameter choices (decomposition, ARIMA model, outliers, seasonal and trend moving averages) are made automatically by X13ARIMA-SEATS. The outcome from the default run is then refined with the over-riding aim being to fit the simplest appropriate adjustment. The end result is then compared with the choices made in the previous review. A decision to alter previous recommendations, or to introduce complications, must be supported by evidence and reasonable argument. User-defined files for prior adjustments (rmx and ppp files) from the previous review are tested for significance and updated where necessary e.g. if transformation type for the series has changed.

This robust approach is taken to avoid uninformative revisions caused by minor changes to seasonal adjustment settings, changes that could easily revert back in the next review.

Findings of the review

The recommended seasonal adjustment models are shown in Figure 2. Two of the series have revised models (Intermediate and Investment) and there were some changes to the parameters for several of the other series. As a result of the impact of the COVID-19 pandemic, many of the series continue to have additive outliers applied in 2020, particularly in Quarter 2. Further level shifts and in some cases additive outliers have been added to some of the models. Additive outliers and calendar effects have also been removed from some of the models.

Figure 2 Seasonal Adjustment Model changes
Series Current Transform Current Model Current TMA Current SMA Current Regressors Revised Transform Revised Model Revised TMA Revised SMA Revised Regressors
IOPLog(0 1 0)(0 1 1)5(3x5)Rp2008.3-2009.2, LS2017.2, AO2017.3 & AO2020.2no changes----
ALLCLog(0 1 1)(0 1 1)5(3x9)Easter[8], QI2008.2-2008.4, LS2017.2, AO2017.3 & AO2020.2Log(0 1 1)(0 1 1)5(3x9)AO2005.4, QI2008.2-2008.4, LS2017.2, AO2017.3 & AO2020.2
ELog(0 1 3)(0 1 1)5(3x5)AO2012.1, AO2013.1, AO2020.1, AO2020.2, AO2020.3 & AO2021.1Log(0 1 3)(0 1 1)5(3x5)AO2012.1, AO2013.1, AO2020.1, AO2020.2, AO2020.3
ALLDLog(1 0 1)(0 1 1)5(3x3)AO2005.2, LS2008.1, TC2020.2, AO2021.4 & AO2022.1no changes----
BNone(1 0 2)(0 1 1)5(3x5)LS2021.1 & LS2023.3None(1 0 2)(0 1 1)5(3x5)LS2021.1, AO2020.2 & LS2023.3
ENGLog(0 1 2)(0 1 1)5(3x3)AO2020.2no changes----
TOTOTHERLog(0 1 2)(1 1 2)5(3x5)Easter[8], td1coef, Rp2008.2-2009.1, AO2013.3, LS2018.4, AO2020.2 & LS2021.1Log(0 1 2)(1 1 2)5(3x5)td1coef, Rp2008.2-2009.1, LS2018.4, AO2020.2 & LS2021.1
CALog(0 1 1)(0 1 1)7(3x9)AO2005.3, AO2005.4, Rp2017.1-2017.3 & AO2020.2no changes----
CHLog(0 1 1)(0 1 1)5(3x9)Rp2008.2-2008.4 & AO2020.2no changes----
CECFNoadjustmentneeded--Noadjustmentneeded--
CBLog(0 1 1)(0 1 1)7(3x9)LS2019.1 & AO2020.3Log(0 1 1)(0 1 1)7(3x9)AIC TEST(Easter, td), LS2019.1, AO2020.3
ConsumerLog(0 1 1)(0 1 1)5(3x5)AO2005.3, AO2005.4, Rp2017.1-2017.3 & AO2020.2Log(0 1 1)(0 1 1)5(3x5)AO2005.3, AO2005.4, Rp2017.1-2017.3 & AO2020.2, AO2025.1
IntermediateLog(0 1 2)(0 1 1)5(3x5)Easter[15], Rp2008.3-2009.2, AO2020.2 & LS2022.2None(0 1 3)(1 1 2)5(3x5)Rp2008.3-2009.2, AO2020.2 & LS2022.2
InvestmentNone(1 1 0)(0 1 1)5(3x5)AO2020.2None(0 1 0)(0 1 1)5(3x5)AO2020.2, LS2009.1
CCLog(0 1 1)(0 1 1)5(3x5)Rp2008.2-2009.1, Rp2020.4-2021.2 & AO2020.2Log(0 1 1)(0 1 1)5(3x5)Rp2008.2-2009.1 LS2015.4 Rp2020.4-2021.2 AO2020.2
CDCMLog(2 1 2)(0 1 1)5(3x9)LS2016.3 & AO2020.2no changes----
CGLog(1 0 0)(0 1 1)5(3x5)Rp2008.2-2009.1 & AO2020.2no changes----
CICJLog(0 1 2)(0 1 2)5(3x9)LS2009.1, LS2009.2, LS2012.4, LS2015.4, AO2020.2 & AO2020.4Log(0 1 2)(0 1 2)5(3x9)LS2006.1, LS2009.1, LS2009.2, AO2011.2, LS2012.4, LS2015.4, AO2020.2 & AO2020.4
CKNone(0 1 1)(0 1 1)5(3x5)Rp2008.2-2009.1 & AO2020.2no changes----
CLNone(0 1 1)(0 1 1)5(3x9)LS2020.2 & AO2020.3no changes----

TMA = Trend Moving Average
SMA = Seasonal Moving Average

An additive outlier (AO) is a data point which falls out of the general pattern of the trend and seasonal component. Although an outlier may be caused by a random effect, i.e. an extreme irregular point, it may have an identifiable cause such as a strike, bad weather or a pandemic.

A level shift (LS) is an abrupt but sustained change in the underlying level of the time series. The annual seasonal pattern is not changed by a level shift. A temporary change (TC) allows for an abrupt increase or decrease in the level of the series, with an exponentially rapid return to its previous level.

A ramp (RP) is a type of outlier used when a trend is changing too quickly to be considered a natural movement of the trend itself yet is not an instant change where a LS would be a better option. An increasing quadratic ramp (QI) or decreasing quadratic ramp (QD) is used when the rate of change is not constant during the ramp phase. The decision to use ramps is usually based on whether a sharp change in trend level causes problems for the quality of the seasonal adjustment.

Some of the models also include trading day (td) and/or Easter effects.

Impact of the review


Figure 3 below shows the absolute difference between the current seasonal adjustment model applied to the data and the revised seasonal adjustment model applied to the data expressed as a proportion, such that:

Absolute Revision = |yT – yt|/yt where yT = value from the revised model and yt = value from the current model.

The data changes from these revised models are reflected predominantly in shifting the level of the series while the patterns are generally preserved. Figure 3 shows that the impact of the revisions is small in all the reviewed series. The graphical comparisons can be seen in the Annex.

Figure 3 Series Absolute Revisions, Reviewed April 2026
Series Full Span Mean Last 3 Years Mean Final Year Mean Latest Data Point
IOP*0.0000.0000.0000.000
ALLC0.0020.0020.0020.001
E0.0020.0040.0030.007
ALLD*0.0000.0000.0000.000
B0.0030.0080.0080.005
ENG*0.0000.0000.0000.000
TOTOTHER0.0040.0040.0050.004
CA*0.0000.0000.0000.000
CH*0.0000.0000.0000.000
CECF*0.0000.0000.0000.000
CB0.0000.0000.0000.000
Consumer0.0010.0060.0080.006
Intermediate0.0040.0020.0030.000
Investment0.0050.0010.0010.001
CC0.0020.0000.0000.000
CDCM*0.0000.0000.0000.000
CG*0.0000.0000.0000.000
CICJ0.0070.0020.0030.001
CK*0.0000.0000.0000.000
CL*0.0000.0000.0000.000

*No change made to existing model

Review implementation

The revised seasonal adjustment models and parameters will be introduced in the IOP Q1 2026 publication. The seasonal adjustment models and parameters will continue to be reviewed regularly, with the starting point for subsequent reviews being these revised seasonal adjustment models.

Revisions to the seasonally adjusted estimates will be made in accordance with the IOP published policy on revisions, informed by the ESS Guidelines on Seasonal Adjustment.

Annex: Seasonal adjustment time series comparisons

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Figures are indexed to 2023=100

Contact Details

Published by: Economic and Labour Market Statistics Branch

Lead Statistician: Stephanie Bruce

Email: economicstats@nisra.gov.uk

Accessibility contact

Please contact Dissemination Branch for assistance with accessibility requirements or alternative formats. Contact details are:

Email:

Telephone: +44 (0)300 200 7836

Dissemination Branch
NISRA
Colby House
Stranmillis Court
BELFAST
BT9 5RR