Introduction
This report summarises the findings and impact of the latest seasonal adjustment review of the Northern Ireland Quarterly Construction Output Statistics. These Accredited Official Statistics provide a timely measure of the volume and value of construction output in Northern Ireland.
Northern Ireland Construction Output Statistics are produced using data from the Quarterly Construction Enquiry (QCE). The QCE is a sample survey carried out by NISRA as part of the Quarterly Business Survey (QBS). More information on the QCE and the Quarterly Construction Output Statistics can be found on the Construction Output Statistics webpage.
A seasonal adjustment review was carried out in May 2026 by the QCE team within NISRA in conjunction with the Office for National Statistics (ONS). The aim of the review was to ensure that seasonal adjustment for the QCE series remains appropriate and is working well. Prior to this review, the seasonal adjustment models used by NISRA were determined from an ONS review carried out in May 2025.
Background
Construction output can be affected by events throughout the year given that some work may be seasonal (for example, firms may have reduced capacity during holiday periods, there may be increased demand at certain times of year, and construction activity may be affected by the weather). The estimates published in the Quarterly Construction Bulletin are seasonally adjusted to account for such seasonal trends. Over time these trends can change, so seasonal adjustment models are periodically reviewed.
Review Methodology
The 20 quarterly series reviewed are shown in Table 1. The 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.
| Series | Description |
|---|---|
| AW | All Work |
| ANW | All New Work |
| ARM | All Repair and Maintenance |
| IH | Index of Housing |
| II | Index of Infrastructure |
| IOW | Index of Other Work |
| NHPR | New Housing: Private |
| NHPU | New Housing: Public |
| NIPR | New Infrastructure: Private |
| NIPU | New Infrastructure: Public |
| ONWP | Other New Work: Public |
| ONWPR | Other New Work: Private |
| PRI | All Work: Private |
| PUB | All Work: Public |
| RMHPR | Repair and Maintenance Housing: Private |
| RMHPU) | Repair and Maintenance Housing: Public |
| RMIPR | Repair and Maintenance Infrastructure: Private |
| RMIPU | Repair and Maintenance Infrastructure: Public |
| RMOWPR | Repair and Maintenance Other Work: Private |
| RMOWPU | Repair and Maintenance Other Work: Public |
Any exact additive relations that hold between series before
seasonal adjustment are not guaranteed to be preserved between the
seasonally adjusted series. However, such relations will still hold
approximately.
From the date of the last review, there have been revisions to the unadjusted data that were previously reviewed due to updates to the construction output indices (OPIs) and an index rebase to 2023. While these revisions affect the full span of the series, the patterns are generally preserved. Further details on revisions to Construction Output Statistics can be found in our revisions policy. In addition, another four quarters of data (Q1 to Q4 2025) have been added to each series.
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 2000 to Quarter 4 2025. Each review included:
assessment of whether the series is seasonal
choosing the appropriate decomposition type, that is additive or multiplicative
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
deciding the lengths of the seasonal and Henderson trend moving averages
reviewing X-13ARIMA-SEATS diagnostics, both quantitative and visual
plotting original and seasonally adjusted series
For series common with the previous review, the old parameters were re-assessed and changed where appropriate.
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 X-13ARIMA-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 were 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 Table 2. Seasonal adjustment is recommended for 15 of the reviewed series, two of which (IOW and NHPU) are not currently adjusted. Five of the series (IH, NIPU, ONWP, RMHPU and RMOWPU) have revised models and there have been changes to the parameters for several of the other series.
| Series | Current Transform | Current Model | Current TMA [Note 1] | Current SMA [Note 2] | Current Regressors | Revised Transform | Revised Model | Revised TMA [Note 1] | Revised SMA [Note 2] | Revised Regressors |
|---|---|---|---|---|---|---|---|---|---|---|
| AW | No | seasonal | adjustment | – | – | No changes | – | – | – | – |
| ANW | No | seasonal | adjustment | – | – | No changes | – | – | – | – |
| ARM | Log | (2 1 0) (0 1 1) | 5 | 3x5 | AO2001.1, LS2013.1, AO2020.2 | No changes | – | – | – | – |
| IH | None | (0 1 1) (0 1 1) | 5 | 3x5 | AO2000.2, LS2001.1, AO2005.4, AO2018.4, AO2020.2, AO2020.3 | None | (1 1 1) (0 1 1) | 5 | 3x5 | AO2000.2, LS2001.1, AO2005.4, AO2018.4, AO2020.2, AO2020.3 |
| II | None | (2 1 2) (0 1 1) | 5 | 3x5 | AO2005.1, AO2012.1, AO2019.3, AO2020.2, LS2020.4 | None | (2 1 2) (0 1 1) | 5 | 3x5 | AO2012.1, AO2019.3, AO2020.2 |
| IOW | No | seasonal | adjustment | – | – | None | (1 1 0) (0 1 1) | 5 | 3x5 | AO2001.1, AO2020.2 |
| NHPR | Log | (2 1 2) (0 1 1) | 5 | 3x5 | LS2009.3, TC2020.2 | Log | (2 1 2) (0 1 1) | 5 | 3x5 | td, LS2009.3, TC2020.2 |
| NHPU | No | seasonal | adjustment | – | – | None | (0 1 1) (0 1 1) | 5 | 3x5 | LS2011.1, LS2013.1 |
| NIPR | No | seasonal | adjustment | – | – | No changes | – | – | – | – |
| NIPU | Log | (0 1 1) (0 1 1) | 5 | 3x5 | AO2000.4, LS2013.1 | Log | (0 1 3) (1 1 2) | 5 | 3x5 | AO2000.4, LS2013.1 |
| ONWP | Log | (0 1 1) (0 1 1) | 5 | 3x5 | AO2020.2 | Log | (3 0 3) (2 1 3) | 5 | 3x5 | AO2020.2 |
| ONWPR | Log | (0 1 1) (0 1 1) | 5 | 3x5 | TC2020.2 | Log | (0 1 1) (0 1 1) | 5 | 3x5 | AO2020.2 |
| PRI | None | (0 1 1) (0 1 1) | 5 | 3x5 | TC2001.1, TC2020.2 | No changes | – | – | – | – |
| PUB | None | (0 1 1) (0 1 1) | 5 | 3x5 | AO2000.4, LS2013.1, AO2020.2 | No changes | – | – | – | – |
| RMHPR | None | (0 1 1) (0 1 1) | 7 | 3x9 | AO2003.3, AO2005.1, AO2018.2, AO2018.3, AO2020.2, LS2024.3 | None | (0 1 1) (0 1 1) | 7 | 3x9 | AO2005.1, AO2018.2, AO2018.3, AO2020.2 |
| RMHPU | None | (2 1 0) (0 1 1) | 5 | 3x5 | LS2001.3, AO2001.4, LS2010.2, LS2013.1, AO2017.3, LS2022.4 | None | (2 1 1) (1 1 1) | 5 | 3x5 | LS2001.3, AO2001.4, LS2010.2, LS2013.1, AO2017.3, LS2022.4 |
| RMIPR | No | seasonal | adjustment | – | – | No changes | – | – | – | – |
| RMIPU | None | (0 1 2) (0 1 1) | 5 | 3x5 | LS2007.1, AO2015.4, LS2020.4 | No changes | – | – | – | – |
| RMOWPR | No | seasonal | adjustment | – | – | No changes | – | – | – | – |
| RMOWPU | Log | (0 1 2) (0 1 1) | 5 | 3x5 | LS2006.4, LS2007.3, LS2008.1, LS2018.1, AO2024.1 | Log | (1 2 2) (1 1 1) | 5 | 3x5 | LS2006.4, LS2007.3, LS2008.1, LS2018.1, AO2024.1 |
[Note 1] TMA (Trend Moving Average) = Length of Henderson Filter
[Note 2] SMA (Seasonal Moving Average) = Order of 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.
Some of the models also include trading day (td) effects.
Impact of the review
Table 3 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.
Overall, the revisions are not large, meaning that the seasonal adjustment is quite stable. Graphical comparisons can be seen in the Annex.
| Series | Full Span Mean | Last 3 Years Mean | Final Year Mean | Latest Data Point |
|---|---|---|---|---|
| AW | 0.000 | 0.000 | 0.000 | 0.000 |
| ANW | 0.000 | 0.000 | 0.000 | 0.000 |
| ARM | 0.000 | 0.000 | 0.000 | 0.000 |
| IH | 0.001 | 0.002 | 0.002 | 0.003 |
| II | 0.008 | 0.004 | 0.003 | 0.000 |
| IOW | 0.012 | 0.009 | 0.007 | 0.001 |
| NHPU | 0.037 | 0.034 | 0.046 | 0.011 |
| NHPR | 0.016 | 0.019 | 0.017 | 0.004 |
| NIPU | 0.001 | 0.003 | 0.005 | 0.004 |
| NIPR | 0.000 | 0.000 | 0.000 | 0.000 |
| ONWP | 0.001 | 0.002 | 0.003 | 0.004 |
| ONWPR | 0.000 | 0.000 | 0.000 | 0.000 |
| PUB | 0.000 | 0.000 | 0.000 | 0.000 |
| PRI | 0.000 | 0.000 | 0.000 | 0.000 |
| RMHPR | 0.007 | 0.007 | 0.007 | 0.009 |
| RMHPU | 0.002 | 0.001 | 0.001 | 0.001 |
| RMIPR | 0.000 | 0.000 | 0.000 | 0.000 |
| RMIPU | 0.000 | 0.000 | 0.000 | 0.000 |
| RMOWPR | 0.000 | 0.000 | 0.000 | 0.000 |
| RMOWPU | 0.001 | 0.003 | 0.004 | 0.005 |
Review Implementation
The revised seasonal adjustment models were introduced in the Quarter 1 2026 publication. The seasonal adjustment models and parameters will continue to be reviewed annually, with the starting point for subsequent reviews being these revised models.
Revisions to the seasonally adjusted estimates will be made in accordance with the QCE published policy on revisions, informed by the ESS Guidelines on Seasonal Adjustment.
Annex: Seasonal adjustment time series comparison graphs
Figure 1: All Repair and Maintenance (ARM), no changes
Figure 2: Index of Housing (IH)
Figure 3: Index of Infrastructure (II)
Figure 4: Index of Other Work (IOW), not previously
adjusted
Figure 5: New Housing Private (NHPR)
Figure 6: New Housing Public (NHPU), not previously
adjusted
Figure 7: New Infrastructure Public (NIPU)
Figure 8: Other New Work Public (ONWP)
Figure 9: Other New Work Private (ONWPR)
Figure 10: All Work: Private (PRI), no changes
Figure 11: All Work: Public (PUB), no changes
Figure 12: Repair and Maintenance Housing Private (RMHPR)
Figure 13: Repair and Maintenance Housing Public (RMHPU)
Figure 14: Repair and Maintenance Infrastructure Public
(RMIPU), no changes
Figure 15: Repair and Maintenance Other Work Public
(RMOWPU)
Contact Details
Published by: Economic and Labour Market Statistics Branch
Lead Statistician: Cathy White
Email: economicstats@nisra.gov.uk
Accessibility contact
Please contact Dissemination Branch for assistance with accessibility requirements or alternative formats. Contact details are:
Email: info@nisra.gov.uk
Telephone: +44 (0)300 200 7836
Dissemination Branch
NISRA
Colby House
Stranmillis Court
BELFAST
BT9 5RR