RDP 2026-05: Real GDP Forecasts by International Organisations Read Me File

This ‘read me’ file details the supplementary information for RDP 2026-05. The publicly available data used to plot the figures appearing in the RDP can be found in ‘rdp-2026-05-graph-data.xlsx’.

The R code was run on R version 4.5.2. on Windows 11. Since only some of the raw data can be made available in this supplementary information due to third-party provisions, the code can not be run in full. However, the code files have been provided so that our methods can be understood.

Raw_data

This paper uses a dataset comprising four international organisation and the average of forecasts made by respondents in the monthly Consensus Economics survey. (See Section 3 of the paper). This is in the ‘Raw_data’ folder.

The supplementary information includes the forecasts of two forecasters:

• European Commission (EC) – these were all downloaded from the EC's publicly available Annual Macro-economic (AMECO) archive. They are included in the supplementary information.
• Organisation for Economic Co-operation and Development (OECD) – these were all downloaded from the OECD's website, previously from the ‘OECD Stat’ page and more recently from ‘OECD Data Explorer’. They are included in the supplementary information.

The supplementary information also contains two spreadsheets of metadata:

• country_facts.xlsx: data on individual countries, including the time period over which their annual real GDP data is measured.
• Publication_dates.xlsx: the dates that forecast vintages were published.

The supplementary information omits the forecasts of the following:

• International Monetary Fund (IMF) – we use forecasts from the World Economic Outlook (WEO). We use forecasts from both the ‘full’ WEO (from April and October) and the WEO updates (from January and June). Our dataset of WEO forecasts has not been included in the supplementary information as it includes some forecasts that are not public. However, the IMF publishes the WEO forecasts from the full WEOs in April and October on its website.
• World Bank – we use forecasts from the Global Economic Prospects (GEP). Our dataset of GEP forecasts has not been included in the supplementary information as it includes some forecasts that are not public. However, the World Bank publishes its GEP forecasts for country groups and selected countries on its website.
• Consensus – we use the average of the forecasts made by respondents to the Consensus survey. These forecasts are available via a paid subscription from Consensus Economics.

Folder structure

The main replication files have the following structure:

Code

Raw_data

Mid_data

Graph_data

Graphs

[same structure as Graph_data]

Tables

Code

The figures and tables in the paper can all be replicated by running the R files in the ‘Scripts’ sub folder. Each script uses a variety of functions that are defined in the R files in the other sub folders of ‘Code’.

This table describes the scripts. They must be run in order, as earlier scripts often write files to ‘Mid_data’ that are used by later scripts. See ‘Raw data’ section for availability of input/output files. The scripts that produce graphs also write the data underlying those graphs to corresponding files in the Graph_data folder. Figure 16 was produced in Inkscape and is provided in the Inkscape_figures folder.

Mid_data

As shown in the table above, some of the scripts will import raw data, transform it, and save to ‘Mid_data’. Other scripts will import from ‘Mid_data’, transform it, and then write to ‘Mid_data’ or produce graphs or tables.

The supplementary information includes many of the files in ‘Mid_data’ produced by these scripts. However, it omits some files that could be used to back out the forecasts of the IMF, GEP or Consensus, which as discussed under the ‘Raw_data’ section of this readme, have not been made available as part of the supplementary information.

21 September 2026

• Supplementary information Download 280KB

Back to abstract